Would the Real Economy Please Stand Up

More-or-less an exegesis of The Manual Economy. Warning: I am not an economist; this is speculative.


What does “the economy” mean to you? What do those words point to in the world? If I say instead “the banana”, that’s a pretty well-defined concrete object. If I say “the marriage”, that’s way less concrete but still has pretty well-defined boundaries in a lot of ways. But when I say “the economy”… well. Isn’t everything kinda part of the economy? It starts to involve a lot of hand-waving and ambiguity.

Just as the idea of “marriage” is made up of roughly three distinct pieces in reality (a legal piece, a social piece, and a now-optional religious piece), I like to think of the “economy” as really made up of a few fairly distinct pieces. Like two kids in a trench-coat, every economy is really two economies trying to sneak into a movie theatre together: a concrete economy made of goods and services, and a virtual economic coordination mechanism. These two economies usually stay approximately isomorphic to each other; when I buy a banana there are two parallel exchanges, one in money and one in bananas. Think of it like double-entry bookkeeping.

To draw out this distinction, imagine two alternative worlds. Imagine first the world vaguely hinted at in The Manual Economy, the world where money and all other coordination mechanisms have completely disappeared, but life carries on completely normally. In this world people still work, and receive goods. There are still “rich” and “poor” lifestyles, and poverty traps, and all the rest. It’s just that there’s no terrestrial coordination mechanism anymore; the economy stays coordinated because some twisted god wills it to be so. In this world you don’t go buy a banana. You go take a banana from the store without giving anything in return. The twisted god controlling us all merely wills it so that poor people don’t go take all the bananas.

For our second imaginary world, take the mirror image of the first. In this world money, and banks, and mortgages and all the rest still exist. People get paycheques, and try to make their rent each month, and get punished when they’re late paying their bills. But also it’s basically The Matrix; everybody eats and sleeps and lives in a permanent tube with no actual goods or services being exchanged. In our first imaginary world, we had a concrete economy running somehow with no coordination. In our second, we have a coordination mechanism running somehow with no concrete economy to coordinate.

Neither of these imaginary worlds are intended to be realistic or “stable” realities in any sense. They’re designed to provide a sense of what these two economies look like when completely isolated from each other. In reality, they depend on each other to stay standing, just as both kids depend on each other to successfully purchase that movie ticket.


I said before that these two economies, the concrete and the coordination, are generally isomorphic. But they aren’t always isomorphic to each other, and unsurprisingly, all the interesting stuff (bubbles, crashes, government stimulus, most white-collar crime, etc.) happens in the gaps where they diverge.

Let’s start with the dual-economy view of an economic crash. In fact, like “the economy” itself, the dual-economy view implies that there are actually two main types of economic crash, depending on which piece of the economy has actually crashed. The easy one to talk about is a concrete economic crash, which can happen when there’s a shock to actual production somewhere. Say that a new disease destroys all of the food crops in North America in the span of a few weeks; this is a real economic crash because all of a sudden the real economic capacity of the continent has drastically shrunk. The problem isn’t one of coordination; even if you coordinate everything perfectly, there just isn’t enough food.

This sounds bad, but the other kind of crash is in some ways worse. When the coordination mechanism crashes instead, there’s generally still enough real economic capacity for everybody to get what they want, it’s just wildly misallocated. This is what happens when for example speculation drives up the price of a particular good until the bubble finally bursts. That can happen even though production and consumption of the good in question stayed perfectly even throughout the entire run. There were no shifts in real supply or demand to justify the bubble. The bubble wasn’t created by a shortage, and didn’t burst because of a surplus in the market. The coordination mechanism simply failed. This kind of crash is what we’ve seen the most of recently: entire neighbourhoods evicted from their subprime mortgages while the homeless crowd the streets; people going hungry while food rots in a warehouse a few blocks away.

The virtual economy can crash because of a bug in the “software” of the coordination mechanism, without affecting the fundamentals of there being enough real capacity to produce all the food, shelter, goods, services, etc. that people actually want. Conversely, the concrete economy can crash (say because of a global pandemic that forces most real economic activity to grind to a halt) but as long as the software is resilient the coordinating mechanism will keep chugging along, coordinating whatever economic activity is left. It’s arguable whether this is a bug or a feature.


Another interesting place where this dual-economy view comes in handy is getting a possible sense of why national debts are so weird. As far as I can tell, the consensus among professional economists is something along the lines of “you don’t want it to get too high relative to GDP, but an increasing absolute national debt isn’t a problem the way it would be for an individual”. This has always struck me as counter-intuitive, and I think a dual-economy model makes it a little clearer.

Debt is an abstract concept, which puts it firmly into the coordination side of the economy. It’s a promise of real economic activity in the future, which allows economic coordination across time. Importantly, there is no real economic “lack” behind debt; you can’t have a physically negative number of bananas. Debt is always just a number on a balance sheet. Just like personal debt can be a useful tool in the form of a credit card – assuming you hold up “both sides” of the temporal bargain by paying it off promptly in the future – so too can national debt. But whereas a personal debt is tied to your personal ability to pay it off, which is finite and limited, national debt is tied to the theoretically-infinite national future.

Now obviously we don’t expect any country to last forever. There’s the heat death of the universe to worry about after all. But the timescales are so wildly different that it’s hard to reason about. It seems plausibly reasonable to sell a twenty-year mortgage to a thirty-year-old human, even if they’re already in debt, but rather less reasonable to sell it to somebody already pushing one hundred. The earnings potential and lifespan just don’t seem to be there to support the future end of the bargain. But it’s entirely reasonable to believe that major economic powers might continue to exist in some form for hundreds more years, and even continue to grow economically through that time. Today’s national debt is an attempt to coordinate real economic activity across time with our future, much larger, national economic future.

Interestingly, a similar kind of analysis can be applied to fractional-reserve banking (with deposits viewed as individual loans to the bank).


I am not very confident in the above analysis, but even if it’s mostly garbage I think that treating the virtual and the concrete economies as separate entities has helped me. “The economy” has always seemed like a big ball of spaghetti with a bunch of arbitrary rules, and I feel like this is a good step towards drawing out a gears-level understanding, even if a lot of the details are wrong.

Hopefully it helps you, too.

Changes in Reality

[Some short thoughts I just wanted to get out of my brain; bullet-points instead of well-structured prose. This is entirely random speculation.]

  • Social systems (laws, customs, memes) are subject to evolutionary pressure from the dynamics of reality; when reality changes, existing social systems are typically no longer in equilibrium and have to evolve, or collapse and be rebuilt. Consider for example the invention of the birth control pill and the resulting impact on family structure, gender relations, etc. Pre-pill social customs around marriage and family were no longer in equilibrium in a world with reliable female birth control, and so society shifted to a new set of customs.
  • “Change in reality” largely means economic and technological change. New wealth and new capabilities.
  • “Change in reality” has been accelerating for a long time as new technologies and discoveries unlock new economic prosperity which enables more discoveries, in an explosive feedback loop. Some argue that technology/science have slowed down a lot recently, but I think that’s mostly because our best and brightest are too busy extracting economic value from our recent innovations (computers and, separately, the internet). Once that bounty has been consumed, more general technological progress will resume its previous course.
  • There is a natural limit on how fast social systems can evolve. Humans can adapt to living under radically different memeplexes, but not instantly, and somebody has to invent those memes first. When reality changes slowly this is fine, as it leaves plenty of time for a multiplicity of experimental memetic shifts in different groups, letting the best adaption dominate with high probability.
  • At some point in the future (possibly soon?) reality will start changing faster than our social systems can adapt. Our existing laws, customs, memes, and government will be out of equilibrium, but we will not have enough time to converge on a new social system before reality changes again. Society will fragment and human culture will undergo an intense period of adaptive radiation.
  • The countervailing force is technology’s ability to connect us (the “global village”) and equivalently the law of cultural proximity.

The Manual Economy

[An attempt at fiction in the style of Scott Alexander. With bits of Lewis Carroll and Douglas Adams thrown in for good measure.]

The hallucination started out so normally, I completely forgot that I was tripping.

I was at the dentist, and I had just had my teeth cleaned. You know the drill, the hygienist goes through your teeth with this little spray nozzle that gets into all the cracks and cavities you pretend don’t exist when you’ve got a brush in there. Then they make you hold some disgusting not-quite-mint not-quite-water in your mouth, and swish, and spit. And spit. And spit. And after about the third blessed mouthful of real water, you can vaguely taste something other than not-quite-mint, until your salivary glands give up the ghost entirely and your mouth turns into the Sahara desert.

As I said, it was weirdly normal for a trip. I’d been expecting unicorns, or aliens, or a sky made up of funky colours and mystical cactus people who could factor large numbers. But I was at the dentist. If I’d wanted a trip to the dentist, I would have just gone to the dentist. It would have been cheaper, and probably better for my teeth.

The entire dental experience was so totally normal I completely forgot I was tripping until I went to pay, and I couldn’t find my credit card. Or any cash. My wallet had a driver’s license and various other identification cards, but no payment at all. The receptionist smiled at me politely.

“Is everything alright? Can I help you”?

I winced. “I’m sorry, I seem to have misplaced all my money, I’m not going to be able to pay my bill today”.

There was a confused pause. A giant hand walked past waving an umbrella and whistling show tunes. The receptionist winked at me with both eyes at once. I suddenly knew, somehow, that I didn’t need to pay, so I turned and walked out the door. Across the street was a bank, so I floated forward until I was inside.

The bank, like the dentist, seemed totally normal. There were no lines, but that was expected for mid-afternoon on a Tuesday. I rolled over to one of the tellers.

“Excuse me”, I said, “I seem to have lost my credit card, can you help me”?

There was another confused pause. The bank teller turned into a giant hand and flew away. The entire bank building sort of dissolved as the buildings on either side squeezed together to fill up the space. I ended up on the sidewalk outside a Starbucks.

I didn’t even like Starbucks.

Sitting outside the Starbucks was a homeless person whose baseball cap kept flickering as if it couldn’t make up its mind. First it was on their head, but then *pop*, it was on the sidewalk in front of them with a few coins in it, and then *pop*, it was gone entirely. And then it was suddenly on their head again. After a few seconds of this my own head started to hurt, so I stared at the sidewalk extra hard until the homeless person turned into a giant hand, and the baseball cap was arrested for multiplying entities beyond necessity.

The hand spoke to me. “Now look what you’ve done! It’s hard enough to coordinate this economy without some yokel trying to physically instantiate all of the mechanisms”!

There was a third confused pause, but this time the hand just sat there looking disgruntled until I finally echoed its statement back as a question. “You… coordinate the entire economy”?

“Yes of course I do”, the hand replied, “somebody has to do it or this whole place would fall apart. How else does food get to everyone who needs it, let alone all the other goods and services”!

I blinked. “So, you’re, literally, the invisible hand of the market”?

“Well I was“, the hand said waspishly. “But do I look invisible to you”?

“Oh, sorry about that”, I apologized. “So my money and credit card, and the bank and everything? They all disappeared because they’re… you? Or manifestations of you, or something”?

The hand glared at me. “I’m a hand”, it said, waving at itself sarcastically. “It seems awfully rude of you to talk to me about manifestations. Until you came along, I had no need of them at all”! It huffed. “Now here I am, trying to coordinate an economy the size of a planet, and instead of being a magical omniscient force I’m trapped in a giant disembodied appendage. What am even I supposed to do with all of these fingers”?

I giggled. “I dunno, you could say that the economy just went… digital“.

The hand rolled its eyes, but I had a lot more ready.

“Oh come on, you’ve got to hand me that one. No? You’re not going to clap back? Well come on then, let me give you a hand coming up with a response. I’m pretty handy with this sort of thing, in fact…”

Ten minutes later, I finally ran out of steam with a complicated pun about greased palms and coconut oil, that even I admitted was a stretch. At this point, the hand had finally had enough.

“Look”, it said, “maybe in your universe the economy is coordinated by these magical distributed pieces of paper and electronic numbers, and nobody has ultimate responsibility for the economy. But in this universe, none of that exists; the buck stops with me. I’ve been listening to you make hand puns for ten minutes, and in that time the entire economy has ground to a halt because I haven’t been there to ensure the right transactions occur at the right time. In some sense I don’t just coordinate the economy – I am the economy”.

I shook my head. “That can’t be right”, I said, “the economy isn’t made up of pieces of paper and numbers, the economy is all of the real things that get moved around because of that coordination. Just because you took a few minutes off to…”, I giggled again, “manifest, as a giant hand, farms are still growing food, factories are still producing goods, the economy is still going! Transport truck drivers didn’t all go on strike because you took a small break”!

“That’s exactly my point!” said the hand. “Truck drivers were on strike when you started your little game, but that strike required coordinated action which I provided. When I started slacking off all those truck drivers got bored and left the picket line to follow their individual inclinations, and now it’s chaos”!

At this point I could feel the drugs starting to wear off, but the hand was still going.

“They’re not striking, or trucking, or anything useful at all! The entire economy is crumbling like the twin towers after that so-called plane crash!”

The bank reappeared beside the Starbucks, and the entire row of buildings shifted down to accommodate.

“It’s all the governments fault, them and their secret mind control beams out to steal your thoughts!”

The not-invisible giant hand shrank in size until it was a normal hand, attached to a normal homeless person, still talking about the implications of omniscient economic coordination and various other conspiracy theories. My teeth started to hurt.

I’ve written this trip report in an attempt to jog my own memory. Something that the hand said during our conversation really resonated with me, and I just know the next Nobel prize in economics is mine if I can only remember what it was…

I just can’t put my finger on it.

Self-Predicting Markets

The story of Hertz has fundamentally changed how I view the stock market. This isn’t a novel revelation – now that I understand it, I’ve seen the core insight mentioned elsewhere – but it took a concrete example to really drive the point home.

The short version of the Hertz story is that the company went bankrupt. They have nearly $20 billion in debt, and as far as anybody can tell, no path to recovery; they’re bankrupt because their business model has been losing money for years despite several attempts to turn things around. The twist? Their stock is still trading, and at time of writing they have a market cap of $900 million.

I notice I am confused.

On any intuitive understanding of the market this shouldn’t be possible. The company is literally worthless. Or really, worse – it’s less than worthless given its debt load. People are paying positive money to own negative money. On a naive view this is another nail in the coffin of the Efficient Market Hypothesis.

After noticing that I was confused, I tried to generate hypotheses to explain this story:

  • Maybe the EMH really is wrong and the markets are nonsense.
  • Maybe bankruptcy laws are so complex and tangled that the expected value of the company really is positive after all is said and done.
  • Maybe the markets expect Hertz to get a government bailout for some reason.

Some of these are plausible (in particular the second), but none of them were particularly satisfying, so I tried asking myself why I, in a hypothetical world, would buy Hertz stock in this situation. I gave myself the standard answer: because I expected the stock go up in value in the future. Then I realized that this answer has nothing to do with the value of the company.

I had been making the mistake of viewing the stock market as a predictor of company value over the short-to-medium term, but this isn’t true. The stock market is a predictor of itself over the short-to-medium term. If people think the stock will go up tomorrow, then the stock will go up today – it doesn’t matter what the value of the company does at all. The company can be literally worthless, and as the Hertz story proves, people will still buy as long as they think the stock will go up tomorrow.

Now in practice, there are a bunch of traders in the market who trade based on the expected value of the company. As long as these people have a majority or at least a plurality, then everybody else is destined to follow their lead. If the expected value of the company goes up, then the expected value of the stock goes up, as long as enough people are trading based on company value. But in cases like Hertz, the expected value of the company is nothing, so value-based traders exit the market entirely. This leaves only the “shallow” stock-based traders, whose rational move is now to trade based on the expected value of the stock being completely divorced from reality.

The market is really weird.

A Living Wage

A couple of unrelated trends clicked together in my brain this afternoon in an unexpected way and I thought I’d share. In case it needs to be said, this is interesting but wild speculation and not meant to be taken very seriously.

One trend is the gradual dissolution of the nuclear family over the last several decades. Another is the more recent economic anxiety in the era of Brexit and Trump. And finally, we have a push from liberals in the last twenty years to raise the minimum wage to what is known as the “living wage“.

As a single adult with no dependants who keeps a comprehensive budget of my finances, I know both what I spend each month, and roughly what it would be possible to live on if times suddenly became tight for me. Interestingly, a full-time job at the minimum wage where I live would be more than enough for a pretty decent single life, since until recently I was living below that line more-or-less by accident. However, it’s also clearly not enough for a family, nor even necessarily enough for a family with two income-earners at that level given the costs of transportation and child-care.

My core speculation is as follows: what if the gradual dissolution of the nuclear family has increased the “supply” of single workers and decreased the “supply” of workers with families, thus driving down the market price for labour and making it even more difficult to have a family. This would be a very unfortunate negative feedback loop, if true.

Maybe somebody with more economic/demographic expertise can dig into this more or tell me where I’m making an obvious error?

Capital is the New American Aristocracy

Do you know that feeling, when some person or article says something with which you’ve agreed for years but hadn’t ever been able to properly articulate?

It’s one of the delusions of our meritocratic class, however, to assume that if our actions are individually blameless, then the sum of our actions will be good for society.

The Atlantic think-piece that this is from (The 9.9 Percent is the New American Aristocracy) is of course excessively long, but worth reading if you’re interested in that kind of thing, especially if you enjoyed my essay on Brexit, Trump, and Capital in the Twenty-First Century. While the Atlantic piece takes a rather different tack to get there, it ends up at roughly the same set of possible recommendations, and I think the underlying thesis is the same: there are a lot of not-directly-monetary ways in which the future prospects of the working class have suffered over the last few decades.

Money may be the measure of wealth, but it is far from the only form of it. Family, friends, social networks, personal health, culture, education, and even location are all ways of being rich, too. These nonfinancial forms of wealth, as it turns out, aren’t simply perks of membership in our aristocracy. They define us.

This was basically one of the theses of my essay, so maybe the Atlantic article doesn’t tack that far from it after all.

Originally, I focused on the value of unskilled labour as the primary change in recent generations. I still think this is generally true, but the Atlantic piece focuses elsewhere: on the walls that the upper classes are building around the other forms of capital. Education is a form of capital, but with the increasing class-segregation of top university admission processes it becomes less accessible to those at the bottom of the heap. Location is also a form of capital, but as the cost of living skyrockets in prime locations (most notable Silicon Valley) that too becomes inaccessible. It seems inconceivable today that a poor person from the slums of Detroit could fight their way into a good university, then afford to move to San Francisco in order to be able to work a decent job. But if we really want social mobility to be a feature of our economy then that’s the story we have to enable.

Finding A Solution

In 2016, I mentioned a basic income as one possible partial solution, with the caveat that it didn’t seem politically or economically feasible. Things have changed a lot in the last two years. From Finland, to California, to Ontario, a number of organizations and governments have started pilot projects of the idea. Perhaps more interestingly, in the current Ontario election, all three major parties have pledged to continue the experiment, effectively guaranteeing a path forward regardless the winner.

In addition to the political will that has sprung up recently, the economics suddenly seem more favourable as well. The cost of expanding Ontario’s experiment to the entirety of Canada has been pegged at only $43 billion, which is eminently affordable in the context of the many hundreds of billions of dollars that Canada already spends on various programs. And proponents are quick to point out that that number doesn’t even include the expected savings in health care, incarceration, and other government services which typically result from lifting people out of poverty.

The purpose of this essay wasn’t originally to sell people on the idea of a basic income, so I’ll leave it at that, but it does seem to me like an extremely promising approach. I’m looking forward to the results of some of the pilot projects, and in the mean time I’m going to do a bit more research and try to raise the profile of this idea.

I hope you’ll join me.


Should Robots Pay Taxes?

I was going to use this as an “other opinions” link but then I started thinking about it and decided to turn it into a proper post instead (my sequence on atheism will resume next Wednesday as usual). Here’s the initial interview: https://qz.com/911968/bill-gates-the-robot-that-takes-your-job-should-pay-taxes/.

It’s an interesting proposal, but it has some weird flaws. For example, how do you define a robot vs. just a tool? Should we be taxing hammers because they let carpenters drive in nails more efficiently, therefore displacing other carpentry jobs? What about one of those fancy smart electric (but still manually controlled) saws? They still seem more like tools, but the line is getting blurry. When you add a CNC module to that saw, does it become a robot for tax purposes? Why? The marginal efficiency of the CNC module itself isn’t necessarily that high.

Another issue is that robots are already getting taxed, albeit indirectly. When a company automates away a job, they do so to save money. That money ends up going somewhere (usually the pockets of executives and shareholders) and tax is payed on it there, usually at fairly high marginal rates. You can argue various counter-points about how much tax those people should fairly pay, but at that point we’ve kind of lost the thread of the argument. It’s not at all obvious that “tax robots” is the right solution to the problem of “rich people are good at tax evasion”, and that wasn’t the original claim anyway.

Of course, if you combine the definitional problems with the tax evasion point then you run into another issue: any reasonable formulation of this tax is going to be trivial to circumvent. You’re going to end up with a single minimum-wage worker pressing a green button once an hour just so the machine at the other end doesn’t meet the definition of a “robot”. I mean sure, your tax has saved a job, so in some sense it’s had the intended effect, but not at the intended scale nor in any way that provides actual quality of life to the person in question.

It’s a neat idea, and it makes for some good headlines, but to me a robot tax just ends up seeming silly.