Adam Ozimek of Modeled Behavior recently chided John Stossel for assuming that the market can handle education. He pointed out that markets aren’t best for everything and gave the provision of security as an example. At an ordinary middle-brow blog I’d led that pass, but I demand better from those that can deliver. I dished out arguments about the remedial state and Bruce Benson’s economic analysis of the provision of law & order in hopes that he’d respond with something clever like Tyler Cowen’s network externalities argument against anarcho-capitalism or Peter Leeson’s against vertically integrated proprietary communities. Now those are people who have spent an unusual amount of time thinking about the issue, but someone as economically literate as Adam could have shot back with a less sculpted version of that type of argument. Instead his first response was rights-based and bereft of public choice (or “New Institutional” if you prefer) analysis, something one might expect from a Randian. It then occurred to me that Bryan Caplan has found a similar phenomena for healthcare. Perfectly intelligent economists will say that it’s so important that usual economic arguments go out the window. To someone who thinks markets are a fantastic way of delivering services, the importance of an industry would suggest all the more reason that it be provided by markets.
Some of this may be the result of being a bullet-biter prone to reductionism and economic imperialism. We ask “what’s so special about X that the usual analysis doesn’t apply”. Robin Hanson is particularly fond of that approach, not only for his specialty in healthcare economics but also asking here why we treat dating and employment differently. An argument for why one should not be a bullet-biter is here.
February 21, 2010 at 6:10 pm
Every country which has some sort of school choice movement–Chile, Canada, the US, Sweden, Netherlands, India, Pakistan, etc.–has shown a substantial improvement.
One might even go so far as to say that the positive impact of education competition is the best established fact in all of education policy! Certainly the effects dominate anything like higher pay for teachers, lower classroom sizes, etc.
February 22, 2010 at 6:23 am
I didn’t criticize Stossel for assuming markets can handle education, I criticized him for making extreme oversimplification that markets can handle everything. Nothing I wrote is a criticism of school choice, which I support.
I’m sure that I haven’t brought the best arguments against privatizing the police. That’s because I haven’t yet seen a reason to think hard about this issue, because it seems so obviously absurd, and the common sense objections are innumerate and sufficient. Pointing out that private police once worked in Iceland in the dark ages doesn’t really help that.
The problem is that, as Milton Friedman once said, your freedom to throw punches gets in the way of my freedom not to be punched in the face- or something therein. Likewise, your freedom to hire people to use violence and coercion against others get’s in the way of my freedom to not be violently coerced. You try and conflate private police with private security, but your private security cannot come to my house, interrogate me, detain me, detain me, and shoot me if I resist. Your private demand for this violent security may be very high, picture the most aggressive police for in the nation operating just barely within the law, and breaking whenever they can do so without being caught. Danger to you is an externality to my safety which I don’t bare when the police work for me. Thus violence of police will be above an efficient equilibrium.
Lastly, let me use one of the examples which you could spend all day coming up with that make this idea seem absurd. A man beats his wife. She calls her favorite police, and he calls his and says she’s beating him. Her police have incentive to use more violence against him, and find him in the wrong, and his police vice versa, since those are their customers. The police are forces are likely differentiated by competition as well, so once force is known as very harsh against woman beaters, and one known as very lax. How does this situation possibly get resolved? It looks like it’s headed for a shoot out, or both man and woman being arrested.
February 22, 2010 at 2:17 pm
“Lastly, let me use one of the examples which you could spend all day coming up with that make this idea seem absurd. A man beats his wife. She calls her favorite police, and he calls his and says she’s beating him. Her police have incentive to use more violence against him, and find him in the wrong, and his police vice versa, since those are their customers. The police are forces are likely differentiated by competition as well, so once force is known as very harsh against woman beaters, and one known as very lax. How does this situation possibly get resolved? It looks like it’s headed for a shoot out, or both man and woman being arrested.”
Wow, he recreated the Ayn Rand shootout scenario. What happens when someone commits a crime in one country that is punished more harshly in the other country that the victim happens to be from. Answer: The two countries immediately go to war killing loads of innocents and wasting massive amounts of blood and treasure…or not.
February 22, 2010 at 5:26 pm
I’m not familiar with the Ayn Rand shootout scenario, but I don’t see how what you’ve sketched is related to my hypothetical. Care to elaborate?
February 23, 2010 at 8:29 pm
Thorfinn, I’m actually not that familiar with studies on the effects of school choice. I’m surprised you include the U.S on that list, since I was under the impression there was actually less of that here than in many European countries.
Regarding the best established fact, I had heard that was the comparative effectiveness of “direct instruction”. Unfortunately, middle class people hate it.
Adam, I suppose I posted in haste and misrepresented you.
Privatizing police is as obviously absurd as letting the market handle shoes. Iceland is a relatively extreme example, but it is not alone. In England before the “bobbies” there were volunteer organizations and professional “thief catchers”, and the introduction of professional police was not all that popular with the populace. Anglo-Saxon clan law was basically all torts, with wergild being the price one paid (to the victims or their relatives) for spilling blood. Admittedly, membership was not based on contract as in Iceland & Ireland, so it might be characterized as more mutualist than anarcho-capitalist. It was gradually over time that kings introduced criminal law, or “crimes against the king’s peace”, in which damages were not awarded to the victim but the king. The Law Merchant operated in a wide variety of jurisdictions, without the need for coercion. It was enforced through ostracism on the part of other merchants. In America, as pioneers headed west in advance of state functionaries, they also set up their own private provisions of law. These and other examples are discussed in Bruce Benson’s “The Enterprise of Law”.
The quote about fists and noses comes from Oliver Wendell Holmes rather than Milton Friedman. Holmes also didn’t think your freedom extended to criticizing the draft during WW1.
Thus violence of police will be above an efficient equilibrium.
It is already above an efficient equilibrium. In a private system of law, police would be held responsible for their acts. If you read Radgeek or Balko, you know that is not currently the case. To use a glaring example, the exclusionary rule prevents illegally obtained evidence from being used in court. But the illegal act itself is never punished, so the rule is of no benefit to someone who has no incriminating evidence.
Domestic violence is problematic under state law already. It’s the kind of call cops least like responding to. I acknowledge that some cases currently resulting in punishment would not under the Hansonian ideal of giving people what they want. It is fortunately the case that married couples have already agreed to a contract, if we were able to choose our own law we might expect said contracts to handle such situations.
Shootouts are inefficient and it is in the interest of most parties to avoid them. The custom of wergild mentioned earlier is an example of how that problem was dealt with in the past. As Snake implied, our existing governments do not leap to war on such bases (though they may for other reasons). International law or the “laws of nations” long pre-date the U.N, and even today to a great extent governments are in a state of anarchy with respect to one another.
A protection (or dispute resolution) agency which behaves in a reckless/violent manner is not going to receive much reciprocity from other agencies. If it seems dangerous enough, it may arouse enough enmity that the other agencies seek to quash it. If it is powerful enough to suppress the rest, the result is the status-quo of monopolized law & order.
February 23, 2010 at 8:39 pm
TGGP
I’m a relatively new follower of yours. However, what do you think of Nozick’s contention that starting from “scratch” the equilibrium of competing protective agencies will end up in Minarchism? Apologies, if you’ve addressed this elsewhere.
February 24, 2010 at 3:24 pm
If I may attempt to boil down your points:
1) There were institutions for crime prevention and justice in places before police as we know them replaced them
2) There are inefficiencies in the current systems
3) If any private security is overly reckless/violent, other agencies will at least not cooperate, and may even quash them.
You haven’t disagreed that a private system would be inefficiently violent in the way I described, it’s unclear why society would want to replace the currently inefficient system with an extremely radical system that is also inefficient by design, but also has never been seen in a modern industrial society (everything you point to is pre-modern and most if not all seem more like different communal systems rather than private for profit police).
You say shootouts are inefficient, but then say that:
“If it seems dangerous enough, it may arouse enough enmity that the other agencies seek to quash it. If it is powerful enough to suppress the rest, the result is the status-quo of monopolized law & order.”
That sounds an awful lot like a lot of shootouts and all out private police war. It’s also not at all obvious that shootouts are inefficient. If private police war creates a state of chaos and spread violence, that greatly increase the demand for private police.
I do enjoy Radley Balko, and I think he points out often changes that could be made along the margin to move our system towards a more just and efficient one. Because violence in my defense has a cost to me that I don’t bare, private police would be inefficient in a deep way that you could not correct at the margins.
February 24, 2010 at 9:07 pm
I should clarify, when I say shootouts aren’t inefficient, I mean from the prospective of the private police it may be optimal for them to engage in.
February 25, 2010 at 1:47 pm
Milwaukee, DC, and New York have various voucher and charter school experiments that have had success.
You’re right about Direct Instruction–though it hasn’t been tested overseas as far as I know. Teachers absolutely hate it as well.
I actually find it frustrating that conservative reformers don’t pick up enough on curricular reform; and puzzling that even private schools have been slow to move in that direction.
February 25, 2010 at 2:00 pm
Thorfinn: Conservatives seem to have devoted more attention to vouchers, though I hadn’t thought it paid off much. Do you remember Bush reading “My Pet Goat” on 9/11? He was actually promoting direct instruction. I don’t think that went anywhere though.
Contemplationist, the possibility (probability, actually) that anarcho-capitalism is not a stable equilibrium is the real reason I am not an anarcho-capitalist, see Randall Holcombe’s Government: Unnecessary But Inevitable. I haven’t read Nozick’s book, Roy Child’s response is that one of his possible scenarios is actually anarcho-capitalism rather than minarchism. Bryan Caplan and Ed Stringham’s paper on networks support’s Child’s contention that such peaceably cooperating agencies are plausible and do not act like a monopolistic cartel.
Adam: By providing evidence that such prior systems existed and persisted for a sizable amount of time I seek to show that the idea is not so “absurd” as you make it out to be. Rather the perception is based on our being steeped in the status-quo and finding it difficult to imagine something else in its stead.
You say that anarcho-capitalism is inefficient by design, I disagree. In Iceland it appears to have been MORE efficient than comparable systems on continental Europe. Rather, we should expect a monopolistic (actually even worse, you can always choose not to buy a monopolist’s good/service) to result in lower consumer surplus.
You bring up industrial society, but have not said why that difference is relevant. My theory to explain the rise of the state is that technology makes it easier for the state to expand its power, but that doesn’t mean citizens benefit from that expansion.
Protection agencies would have more demand during shootouts, but they would also have higher costs. Both agencies and their customers would be better off in the absence of shootouts. Following Ostrom and Coase, I think people are better at coming up with mechanisms to deal with such issues than they are commonly given credit for. I don’t think that’s just a theory, but was historically the case in Iceland relative to Europe.
Violence in defense of you has a cost, the agency used by the person it is inflicted on (which might be the same as your agency doing the inflicting) would be incentivized to ensure that cost is taken into account. A monopolistic state on the other hand can completely ignore the demands of its people and inflict costs on them willy-nilly. See North Korea.
February 26, 2010 at 6:52 am
Far far too much of your argument is based on misapplication of competitive private goods market results. You need to establish that policing is more like a private good than basic and common sense economics suggests.
There are obvious public goods characteristics of policing in many circumstances(local non-rivalrousness, non-exludability) which means you can’t just make the sort of private good assumptions that you assert all over the place (i.e. that costs will be lower, monopolists less efficient, etc.). Compounding that, there are obvious externalities to my security. Waving your hands at these externalities with vague coasian notions and claiming perfect foresight contracts could do away with them makes for an unbelievable argument that in any case.
Your argument that shootouts shouldn’t happen because everyone would be better off without them is contradicted by the persistence of real world shootouts. You say there would be less violence, but then constantly make reference to mechanisms that would have a high probability of being extremely violent:
“Violence in defense of you has a cost, the agency used by the person it is inflicted on (which might be the same as your agency doing the inflicting) would be incentivized to ensure that cost is taken into account.”
So one police force tries to stop you from being arrested, while a competing police force tries to arrest you. Sounds like a shootout. Again, waving your hands at the inefficiency of this doesn’t help, since we constantly see shootouts in the real world. Obviously, given uncertainty and information asymmetry, both sides can and often do believe a priori that it is in their best side to fight.
In either case, the alternative to shootouts is that whoever has the highest willingness is what decides the administration of violence, which is obviously something most societies would reject as acceptable.
And modern industrial societies matter for comparison, by the way, because the cultures, institutions, and entire context is different. You know this, because if someone argues “well, cavemen did it”, it has no bearing on it’s applicability to modern society. It also matters because, as you point out, societies tend to evolve good solutions to these problems, the fact that all societies evolved away from private police should tell you something.
February 27, 2010 at 11:42 am
Policing is excludable (as the practice of outlawry indicates) and rivalrous (poor neighborhoods often complain that rich neighborhoods get more policing).
I never said anything about perfect foresight, I simply think bottom-up actors have FAR better incentives and knowledge to solve such problems. Private law would not give us the war on drugs. The WOD is super-inefficient because the drug users themselves would be willing to pay not to be “served and protected”, and even non-users like myself don’t want to pay for the service at all. Furthermore, the enforcement of drug laws just creates more violence! Peter Moskos claimed that in his time policing Baltimore, basically all the homicides stemmed from the drug trade. Bruce Benson reports that the number of police officers doubled over a period of time without the number of man-hours rising at all. The same sort of problem we see with public education.
“the persistence of real world shootouts”
Those are caused by the illegality of drugs (or other trade). Warring gangs do sometimes try to make a parley (often local ministers assist in this), but it is far more difficult to do so when the police keep getting in the way. Police also pride themselves on netting the “big fish”, the guy on top. But this just results in more violence as underlings fight to fill the vacuum!
“So one police force tries to stop you from being arrested, while a competing police force tries to arrest you.”
This is a forseeable circumstance and so the two agencies will have an incentive to make arrangements for what is to happen in such an outcome. They could have rules for particular situations or arbitration courts who deal with disputes between them.
“Obviously, given uncertainty and information asymmetry, both sides can and often do believe a priori that it is in their best side to fight.”
I’m reading “The Selfish Gene” right now and Dawkins discusses how surprisingly rare deadly fights are in the animal kingdom. There is a lot of blustering and “gloved boxing”. This is because animals have an incentive not to fight, it is costly for both. They develop pecking orders and rules of territory which are evolutionarily stable equilibrium. I think with our larger human capacity for problem-solving, along with contracts and reputation we can do far better.
“And modern industrial societies matter for comparison, by the way, because the cultures, institutions, and entire context is different”
You completely failed to understand the point Robin was making in the linked post. Merely saying “it’s different” is not sufficient, you should explain why that difference (wther culture, institution or “entire context”) is RELEVANT to the problem so that privately provided law is no longer advantageous.
The state does not have its origin in social contract, but in conquest. The fact that it is pervasive does indeed convince me that anarcho-capitalism is not as stable an equilibrium (though it actually lasted a number of centuries in Iceland). But it does NOT tell me that public policing is preferable. I don’t think the Somalis will be able to keep their anarchy, but that fact is unfortunate for them.
March 12, 2010 at 7:30 pm
[...] Are Security and Education Like Health Care? by TGGP [...]