U.S. exchange groups and markets, including NASDAQ and CME Group Inc., will begin trading Bitcoin in 2018. The value of the cryptocurrency has soared to $17,000 a coin in December. (Jhaan Elker/The Washington Post)

Bitcoin changes prices too quickly to be a currency and processes transactions too slowly to be a payments system, but it is juuust right for teaching libertarians everything they don't know about economics.

Not that they're paying attention.

If you listen to bitcoin's biggest backers, it's supposed to be our gleaming future, one where we can make money just by holding it, move it anywhere in the world for free, and no longer have to depend on banks or governments to do the right thing. If you look at what bitcoin actually does, though, it's more like digitized nostalgia for a pre-modern past where money was discovered rather than printed, economics was a simple subject where markets never failed, and you never had to trust anyone you didn't know. It works, then, the way libertarians think things should—which is to say not at all.

The first thing they don't understand is that money isn't just a store of value. It's also a medium of exchange, or what we use to buy things with. And if it's going to be much of one, then it not only has to avoid losing too much value, but also gaining too much. Otherwise, why would you ever spend it? You wouldn't. You'd just hold on to it as long as you could in case, like bitcoin, it went from being able to buy $900 worth of stuff one year to $19,000 the next. Which, if it ever did replace the dollar, would bring the economy to a halt while everyone stopped buying anything other than the essentials and waited to become bitcoin millionaires.

To stop that from happening, you'd need to be able to increase the supply of bitcoins as the demand for them did. This is more or less what is known as “printing money,” and, as is often the case, it can be either good or bad depending on whether it's done appropriately or not. Do it too much and you can get the type of persistent inflation the U.S. had in the 1970s; way too much and the kind of currency-killing hyperinflation Germany had in the 1920s; but too little and the economy might fall into a doom loop like the whole world did in the 1930s. Bitcoin, though, is set up under the assumption that people — or, more accurately, governments — can never be trusted to do this, and that pretty much anything that reduces the value of a currency is by definition bad. That's why its pseudonymous creator decided there would only ever be 21 million coins, even though that hard limit has meant prices have zoomed up and down and back up again as interest in bitcoin has itself. That's made it the best penny stock and the worst currency in the world.

The second thing they don't get is that trust makes economies more and not less efficient. Bitcoin, you see, is best understood as an attempt to rewrite the rules of our money and our financial system so that your savings are safe no matter what happens in Washington or on Wall Street or whatever digital version of them springs up. To make it so nobody has to trust anybody. But it's an ideological point of view that bitcoin takes far beyond any technological need. Here's what I mean by that. The real genius of bitcoin — and there's plenty of it — is that the process of “mining” new coins creates a public record of every transaction it's ever been used for. As a result, you can send things online without needing a bank to tell you who has what to send. That's already there for everyone to see. So goodbye transaction fees, and hello bitcoin!

Well, except for one little thing. The number of transactions bitcoin can process is extremely limited by the fact that it's chosen not to put much memory into its system. Indeed, bitcoin can only handle a maximum of seven transactions a second compared to the 56,000 that Visa can. That means that even though bitcoin's transaction line isn't very long — not many people use it, after all — it still takes a long time to get through it. Unless, that is, you're willing to pay the $28 it now costs to skip to the front. But what's the point of using bitcoin then?

There is, of course, a pretty simple solution here. That's just ... increasing bitcoin's memory. The people who run it, though, have ruled that out. Why? Because that would require a modicum of trust, and they want to abolish that entirely. Bitcoin's raison d'etre, remember, is to reprogram the economy so that governments can't inflate your money away and banks can't gamble it away. Creating a parallel financial system that lets you manage your money outside of the traditional one is the first step in this. Keeping it from becoming as centralized as the old one is the second. “Bigness” in all its forms is the real enemy. It's how you get the kind of single points of failure — the Federal Reserve, Lehman Brothers, or, maybe, even a large enough bitcoin mining group — that can potentially bring the whole thing down. You have to trust that they won't (and regulate them just in case).

If you don't want to do that, then you can't really add more bandwidth to the bitcoin system. Here's why: The more data there is, the more computing power you'd need to win the mathematical races that decide who gets new coins. And in that case, mid-sized miners would have a harder and harder time competing. The market, then, would naturally consolidate into a few big players, and bitcoin's payments system — that's what the miners are really doing — would be just as top-heavy as, say, the credit card companies are today. So just like anything else, specialization would make bitcoin work better, but at the cost of having to trust the specializers. Which, as we've said before, they don't want to do. Bitcoiners would rather keep it pure and useless for anything other than talking about how it must be good for something.

But even in a world where bitcoin actually did work, it still might not be worth using. At least not from a societal perspective. That's because it's not just a matter of how much bitcoins cost people to use, but also how much it costs everyone else when they do — which could be quite a bit. The type of computers that can quickly solve bitcoin's cryptographically complex equations aren't cheap to run. In fact, they're energy hogs.  They already consume more than 0.1 percent of all electricity (or about as much as Denmark), which is remarkable when you consider how little bitcoin is actually used right now. If that ever went up, so would its energy needs — perhaps substantially so. The important thing to understand is that the more bitcoin costs, the more incentive there is to “mine” for it, but the more that happens, the more computing power you need to win new coins. So the amount of energy it uses should go up hand in hand with its price.

Bitcoin, in other words, is one big negative-externality machine. That's what economists call a cost that someone else has to pay for something you did. The canonical example is the pollution that comes out of a factory — society at large is left with the cleanup bill — and bitcoin might not be that different. Sure, some bitcoin miners run on renewable energy sources like hydroelectric or geothermal, but a lot of them still use coal. It's the economical choice, after all. Well, at least for them. So even in the best-case scenario, bitcoin might not be cutting transaction costs so much as redistributing them from individuals to society. That's what it would mean if miners who get paid with new bitcoins replace bankers who get paid with fees as our middlemen.

This, apparently, is progress.

Bitcoin is a revolutionary technology built on reactionary economics. That first part has blinded people to the second — how could something so clever be so useless? — but it's true. Bitcoin's strictly limited money supply harks back to a time when money was a shiny rock you dug out of the ground, not a piece of paper with a dead president (or treasury secretary) on it. And its attempts to insulate miners from the forces of economic rationality are akin to nobles' old feudal protections.

Bitcoin is only the future if you think 1789 wasn't in the past.