Attached is a slightly cleaned-up transcript of an IRC conversation I had last night during which I tried to teach political economics 101 to a well-intentioned person I know who describes himself as a “socialist” (though, from the way he reacted, perhaps not for much longer). It went better than one might have expected.
I think this transcript is interesting on at least two levels. First, it was very constructive, with lessons about how intelligent people who don’t know economics perceive the issues around it as well as a short course in public-choice theory embedded in it. But also…there’s tradition running from Plato’s Dialogues to the Renaissance of teaching philosophy through dialog. It’s interesting, I think, to see how this modulates into the key of IRC.
I am “esr”, of course, and the well-intentioned socialist is “emsenn”. Regular Armed & Dangerous commenter Daniel Franke puts in an appearance as “dfranke”. Also appearing is “Rowan”, who has serious health problems, and “citizen”, who is listening and learning.
This is only very lightly edited. I have fixed typos, removed some extraneous comments by others, reordered a few lines where responses crossed each other, and joined some adjacent IRC lines where that makes it flow better. Material added after the fact is bracketed with , paraphrasing something said in discussion outside the span of the transcript.
emsenn: esr, why does having social programs mean having a heavy hand in the economy?
esr: Because the more of the GDP that goes through political allocation, the more you crowd out market signals. Um, do you know about the “calculation problem”?
emsenn: I’m not knowledgeable in what I’m talking about, if you haven’t noticed. ;) I just know that here sucks, and places with more social services tend to seem better.
emsenn: I’d rather things seem nice and safe than have them be blatantly trashed. I’d rather have our infrastructure seem to be up to date, our streets seem to be in proper repair, our water seem to be drinkable.
esr: (Er, “seem” is the operative word.)
esr: OK, let me explain. Note that I am not making a political argument here; this is value-free economics.
* esr gathers his thoughts.
esr: OK. You have an economy. Some of the resource allocation is through markets. Some of it is through politics. For the moment we’re deliberately ignoring political labels.
esr: The “political” side may be called socialism or communism or just “welfare state”; it doesn’t matter.
esr: The problem is this: in the absence of market [demand] signals, you can’t put capital where it will produce the most benefit to the most people. The information you need to maximize joint utility is only elicited by market transactions.
esr: Hayek noticed in 1938 that even assuming an angelically benevolent political class, this problem is unsolvable.
emsenn: And when the government is heavily involved in economics, it reduces the number of transactions?
esr: Right, but that’s only part of it. [It's not just raw volume of transactions that's significant, but the quality of the demand information in them. Political allocation creates noise - distortion - that degrades that information.]
emsenn: I don’t get why the two are tied. How come government involvement reduces transactions (trade?)
esr: Because capital that would otherwise go into market transactions is getting diverted into political allocation.
esr: But that’s actually not the worst.
esr: The worst is that political allocation quickly gets captured by rent-seekers. So, for example, welfare programs get captured by welfare service providers.
emsenn: Lobbyists, basically?
esr: Yes, and their parasites.
Rowan As far as the social services goes, I’d really, really actually like proper single payer healthcare.
dfranke: Rowan: Health care is vastly overconsumed when consumers don’t have to pay market rate for it, and it ends up making them sicker.
esr: dfranke is correct.
Rowan: Yes. But it’s also vastly unaffordable for people who DO NEED IT.
emsenn nods at dfranke. Isn’t it something like those with health insurance of any sort spend 9x more on health care than those who don’t?
esr: The reason for the capture effect is that the rent-seekers have concentrated incentives to game the system, whereas everyone else has only diffused ones
Rowan: I have disability. I cannot generally afford the healthcare services I need, such as the referral to the pain specialist to see if we can get me proper pain management. I can’t afford the physical therapy twice a week, etc.
dfranke: Rowan: whether in a socialist system or a capitalist one, health care has to be rationed. We’re not at the post-scarcity level for it yet. Capitalism will ration it far more efficiently.
emsenn has no incentives and is still trying to game the system.
emsenn: Well, game it to actually pay attention to what it’s doing and spend money correctly
esr: Yes, but you’re at a competitive disadvantage – you will always lose out to rent-seekers who can hire lawyers and lobbyists.
emsenn: Nuh-uh. I’m friends with their lawyers and lobbyists.
esr: (Note that I’m still not making value-loaded “political” claims. I’m explaining some of the basics of what’s called “public-choice economics”.)
* emsenn nods
Rowan: I have a crap POS healthcare plan right now, and even this I can barely afford.
esr: OK, so you’ll lose out to the people who can hire *more* lawyers and lobbyists than you…
Rowan: Luckily for maybe the next six months, I can afford the out of pocket premiums that I have to pay for the basic things my insurance doesn’t.
Rowan: I can afford new crutches, and thirty bucks for a prescription. But really, I usually can’t.
Rowan: I don’t care if single payer isn’t the system that works best. I just want a system that is affordable and works.
esr: OK, now I can explain why publicly funded health-care is a disaster.
esr: Basically, it means there are no market signals in health care. This leads straight to accelerating malinvestment — too many CAT scanners, not enough primary-care clinics.
Rowan: Right. I can see how that works, yeah.
esr: It locks in a ruinously high rate of cost inflation. The more heavily you subsidize, the worse it gets.
emsenn: esr, what about the fact that your clients, the patients, require these things. Can’t you just take metrics based on that?
esr: Been tried. Look at the NHS in Britain, or the Veteran’s Administration hospitals in the U.S. *shudder* What you end up with is the worst of both worlds; bureacratic allocation and ruinously high costs.
* emsenn nods, fucking VA.
emsenn: (My grandma moved my grandpa out of a private hospice to a VA for some unknown reason, it was horrible)
Rowan: esr: I did say working single payer care. NOOOOOOT the current disasters.
esr: But the theory that any single-payer system can work better is a delusion. The problems are intrinsic and all ground out in the absence of market signals.
emsenn: esr, so what system would work? Please keep in mind that people tend to be stupid, selfish, and mean.
Rowan: Or hell, I’ll be happy if I can get good health insurance if I get a job, but fact is, I’m uninsurable as to most standards. Chronic conditions and a history of complications.
esr: What would work better is a totally free-market system. No subsidies. Without the subsidies, medical price levels would crash – they’re being sustained where they are by the fact that effectively no one in the system is really cost-sensitive.
Rowan: esr: I would definitely support that.
esr: All the medical-services people are, in effect, being told that subsidy programs will support their inefficiencies forever. No wonder they gold-plate everything! You would too!
esr: It’s not that anyone is stupid or evil, it’s that the incentives are all wrong.
emsenn: esr, free market health care would be cheaper? What about, say, IP rights on medication, companies charging huge amounts for helpful medicines that they own sole rights to?
esr: Well, there are several possible solutions to IP blockades. One would be to junk the patent system.
emsenn: So then what’s the incentive for companies to do expensive research if they can’t get insane profits?
esr: emsenn: What is the incentive for computer-chip manufacturers? Drugs are no more IP-intensive than that, and we manage to have a competitive market there. Actually, chip fabs cost more than researching and certifying a drug.
emsenn: Is it? Most chip types have like 2-3 producers
esr: Yes. Keep going, you’re on the right track there…
emsenn: Uhh… the track I’m on is a monopoly on the market.
esr: (Wow. I explain economics and a self-described “socialist” actually listens. This is unprecedented…)
esr: (That was a compliment, emsenn…)
emsenn: esr, I’m only socialist because from what I’ve seen it seems to be best. You’re more than welcome to change my mind. Just don’t expect it to happen in a night
emsenn: Also, omnomnom donuts.
Rowan: Want donuts.
esr: OK, good. You understand that monopolies are a serious problem.
emsenn: They’re like governments without the public accountability :(
esr: I’m not back to politics yet. That involves value commitments like (in my case) “freedom is more important than equality”.
esr: I’m willing to talk in those terms, but I don’t need to right now.
citizen: This is the most mature political online discussion I have seen in a while
emsenn: So, we’ve got 2-3 producers of medicine, all in competition with each other. This is, so far, good. What prevents one from buying the others, or just over time one taking a firm lead over the others?
esr: So, the question of how we avoid market failures almost (not entirely, but close enough for right now) reduces to the question of how we avoid monopolies. You figured that out yourself.
* emsenn gives himself a gold star. Wait no, fuck gold stars, I have donuts. Again, omnomnom.
emsenn: So far we haven’t been very good at stopping monopolies – they tend to end up fucking themselves over.
esr: YES! You’ve noticed that! Good!
emsenn: Mhm. Problem is they tend to set back advancement in the field.
esr: OK, now we got to an empirical question: how is the half-life of monopolies (that is, the time after achieving monopoly status that 50% will crack) related to the amount of political allocation in the economy?
esr: And the answer is….
Rowan: Now we’re beyond Rowan’s understanding of this stuff. So Rowan is going to idle.
* emsenn drumrolls
* emsenn thinks Rowan doesn’t give himself enough credit.
esr: They’re directly correlated. That is, more political allocation -> more persistent monopolies. Freer markets -> less persistent monopolies. Here’s why…
esr: A monopoly is a rent seeeker. A very rich, powerful rent seeker. A monopoly is in the ideal position to buy the regulators!
emsenn: I get what you mean. They can – right. To help ensure ideal conditions for them.
esr: The solution to the monopoly problem, and to market failures in general, is to reduce the amount of political allocation in the system until the monopolies have a short enough half-life to be tolerable.
emsenn: How tolerable is that half life when you’re dealing with lives, as with our health system?
dfranke: emsenn: The question isn’t so much whether it’s tolerable as whether you can do any better.
esr: Thus endeth Public Choice Economics 101.
emsenn nods at esr. Sound logic.
emsenn: And it takes into account the fact that people can be assholes.
Rowan: esr: Next time can we do it in layman’s terms? :) I understood most of that but.
emsenn: So… uh. Explain more sometime. I’m hard-put to find fault with it.
esr: I will. There’s other stuff, like Coase’s Theorem on externalities, that is just as interesting.
* Rowan nods.
dfranke: emsenn: an externality is basically any effect of a transaction, be it positive or negative, that falls on people not involved in it.
esr: Warning, however: If you really get this stuff, to the point where you can do analysis in these terms, you will be at severe risk of turning into a libertarian.
emsenn: esr, I’ve got no problem with that if I agree with it. :P
esr: Coase’s theorem is especially deadly that way.
esr: (Coase’s Theorem is the Killer Joke of political economics.)
emsenn: So like if I paid you to fix my street, an externality of it would be the fact that the people on the street are shit out of luck when it comes to leaving their homes for the duration?
Rowan whimpers. Pain, and an hour until pain pill.
dfranke: emsenn: well, the canonical example is lighthouses.
emsenn: But I want my street fixed :(
dfranke: emsenn: suppose if you build a lighthouse, it creates a benefit far in excess of the cost, and all sailors agree on this.
dfranke: emsenn: but how do you get someone to pay for it?
esr knows where dfranke is going and bows in his direction.
dfranke: emsenn: if you build it, then there’s no way to deprive any sailor of its benefit, whether they paid to help build it or not.
emsenn: Could you tax the ships that go that route?
dfranke: emsenn: this is among the most common rationalizations for coercive taxation.
emsenn: I picked the wrong answer didn’t I?
dfranke: emsenn: Coase’s Theorem, on which I am not an expert and esr will probably have to pick up the slack for me, demonstrates that this ends up not really being a problem; that no matter what resource allocation you start out with, there will be some series of possible transactions that people will consent to that ends up getting you where you want to go.
emsenn: Could tax the nearby town too, I guess, since they get benefit from the trade, which will pick up now that the ships are at less risk of crashing.
dfranke tags esr. Your turn.
emsenn: There’s always a way to get someone to foot the bill even if they aren’t directly benefited, you mean?
esr: Not quite. What Coase’s Theorem actually says is this: “If transaction costs are sufficiently low, all externalities will be internalized” – that is, turned into market transactions between the participants. Almost what emsenn said, but more precise.
emsenn: Hah, cool.
esr: The kicker is that some externalities are hard – pollution, for example. You have to drive transaction costs to infinitesimal or zero before they internalize.
emsenn: Htm…can you look at long term costs? Like “solar costs X more now, but over 20 years the cost is nil compared to spending on coal”.
esr: Right, that’s just net-present-value accounting; it’s easy.
esr: And now you know the real reason the Internet is important.
dfranke: esr: Huh?
dfranke: esr: Oh…reducing transaction costs?
esr: The Internet is the most potent reducer of transaction costs since the invention of money.
dfranke: esr: Hmm, hadn’t thought of it that way before.
dfranke: esr: But you’re self-evidently correct.
esr: So, here’s how you internalize pollution costs in a really free market…
esr: You treat pollution as a form of tortious assault for which individuals can collect damages. Mercury or dioxin in my water table sucks, man!
emsenn: esr, so… treat pollution as an expense, in a way?
esr: Now, you create a futures market in shares of class-action lawsuits.
emsenn: Who will the lawsuits be against?
esr: The polluters.
emsenn: So, everyone? Or just the major polluters?
esr: This only works to the extent you can identify sources of pollution.
esr: The better your sensor technology gets, and the better your ecological modeling is, the lower the threshold of pollution you can internalize.