Bryan Caplan debates Yaron Brook on anarcho-capitalism. You can safely skip Yaron’s parts, but Bryan’s part are golden.
Rothbard.com
Murray Rothbard, Libertarianism, and Anarcho-Capitalism
Bryan Caplan debates Yaron Brook on anarcho-capitalism. You can safely skip Yaron’s parts, but Bryan’s part are golden.
Rothbard.com 
Here is an argument against anarcho-capitalism:
Ancaps support NAP, which states that actions caused by initiated aggression are immoral. By extension, every market that is based on initiated aggression is not a free market. Here is the problem:
In the standard Ancap view, it is the private protection agencies (PDAs) that protect people from aggression. Basically, a private protection market is formed. But the reason why people take part in this market is the credible threat of initiated aggression (i.e one is afraid of being robbed, killed, etc.). But if we review NAP, then this is not a free market, since it exists solely due to the presence of initiated force. In short: Anarcho-capitalism contradicts its own philosophy.
I respect your dedication to rationality.
But, in the ancap framework, force is divided into initiatory force (aggression — always illegitimate) and retaliatory force (defense/enforcement — legitimate). The “Non-Aggression Principle” (NAP) prohibits only initiatory force, not all force.
—
Force: the use of physical power to compel someone against their will. Includes both legitimate (self-defense, enforcement of a just contract) and illegitimate (aggression, robbery) uses.
What you say is correct. Just an addition: The credible threat of initiated aggression is considered “initiatory force” by NAP too. I know, because I have dealt with this topic in depth. But this is exactly the reason people would hire a PDA in an Ancap Society. That is why Anarcho-Capitalism does not conform to NAP ( on which it claims to be based).
I have nothing against the basic idea of Anarcho-Capitalism. It is just the implementation that is totally wrong!
Here is a way to look at the same problem (from one of my articles):
The anarcho-capitalist insurance/defense company solution has a distinct and very specific problem. It is a general one which applies not only to national defense, but to defense from any initiated aggression. It stems from the fact that what is suggested is a fusion between the general economic fields of insurance and police protection. Let us first see how the insurance branch of the economy normally functions. An insurance company typically offers compensation for occurrences such as: natural disasters, loss of productive ability, loss of property, etc. What is common among all of the mentioned cases is that from the point of view of the insurance company, these occurrences are unforeseeable. They can be predicted only through statistical means, but cannot be controlled by the company itself. In short: an insurance company has no control and cannot have any control over earthquakes, for instance. It can only estimate the probability of having one and bet against its clients on its happening/not happening. What the anarcho-capitalist insurance/defense company approach offers is a fusion between insurance and defense. I.e. an insurance company insures against some event and at the same time protects you from it by using its defense branch. What this, however, suggests is that the insurance company will obtain control over the particular insurable event. This way, it will be able to exert control over its happening/probability. Once this control is available, there is no doubt that it will be used in the best interest of that company.
What is typically suggested in the relevant literature is that such control will have a positive outcome for the company’s clients, because the insurance company will try to minimize its costs. For example, if you are insured against robbery, the company will try to protect you, so that it does not have to pay you compensation in case you are robbed. What seems to have been overlooked in this simplistic view is the fact that companies on the market try to maximize their profits in the first place. Minimizing their costs is part of the profit motive, but it is not the whole one. Another part is maximizing their sales revenues. In particular, an insurance company would want to win customers who are highly motivated to insure themselves and thus willing to pay a high price for the company’s service. But what motivates people to insure themselves against aggression in general? This is the fear of being robbed/mugged/killed/etc. If this fear is high, people will be motivated to pay a lot for insurance services. However, if the same fear is low, people will not be willing to pay much to protect themselves from aggression. Thus, from the point of view of the insurance company it is better when its clients are afraid. Their legitimate concerns over their security will translate into bigger insurance sales. People, however, are afraid only if such crimes happen from time to time to them or to their neighbors. From such a standpoint the lack of crime/aggression in a society will be the worst thing that can happen to such an insurance company. If there is no crime, people will not insure themselves against it, so no clients and consequently no revenues for the insurance companies. Note that if one has no clients, one cannot minimize costs in the mentioned way, because these costs exist only if there are clients. All of the above means that it is in the best interest of any insurance company there to be crime/aggression. However, in the anarcho-capitalist solution it is the insurance company which is able to control the occurrence of these offences, since it is given control over them. There is no doubt that this control will be used in the best interest of the company, but this interest may not coincide with the one of its clients. So, crimes may be allowed to happen, because this will be beneficial for the owners of the company.
It must be noted that such a situation can happen because of the fusion of the insurance and executive functions. If such a fusion does not exist, the said problems will not be present. And the reason is that the insurance company will not be able to control the particular insured event the same way it cannot control the occurrence of an earthquake.
From an incentive/motivation point of view, the defense companies have two self-opposing motives: on the one hand, to stop the crime, because only then will they be employed and paid by their clients; on the other hand, they will have motivation not to prevent the crime, because only then will their clients be afraid and therefore willing to pay for their services. This is a classical situation of a conflict of interests.
Sorry for the late reply.
First, it seems we need to review the definition of the Non-Aggression Principle (NAP.) Per Murray Rothbard in “War, Peace, and the State:” “The fundamental axiom of libertarian theory is that no one may threaten or commit violence (”aggress”) against another man’s person or property. Violence may be employed only against the man who commits such violence; that is, only defensively against the aggressive violence of another. In short, no violence may be employed against a non-aggressor. Here is the fundamental rule from which can be deduced the entire corpus of libertarian theory.” This is different from saying that “actions caused by initiated aggression are immoral,” because escaping from a thief is not, nor is calling for help, etc.
Second, the “by extension, every market that is based on initiated aggression is not a free market” claim is ambiguous. It can be a market whose transactions themselves involve initiatory force (e.g., a slave market), which is prohibited by NAP because the exchanges within it require acts of aggression. Alternatively, it can be a market whose demand is driven by the existence of initiatory force in the world (e.g., private defense agencies), which is not, since it responds to the fact that aggression exists while its own transactions remain voluntary.
Third, I’m not sure I follow your train of thought when you say: “But if we review NAP, then this is not a free market, since it exists solely due to the presence of initiated force.” The NAP condemns acts of initiatory force. It does not condemn markets that arise in response to the existence of initiatory force. The relationship between the market and the evil it addresses is responsive, not constitutive.
Fourth, the conclusion that the contradiction exists is, hopefully, clearly wrong as we do not accept the broad definition of “caused by initiated aggression.” In fact, the latter would also render illegitimate any market for self-defense, locks, etc.
Fifth, regarding the principle that “the credible threat of initiated aggression is considered ‘initiatory force’ by NAP too,” it is correct with respect to the person making the threat. The distinction between a “credible threat” and “empty rhetoric” is what justifies a defensive response. To quote Rothbard in “The Ethics of Liberty:” “It is important to insist, however, that the threat of aggression be palpable, immediate, and direct; in short, that it be embodied in the initiation of an overt act.”
Sixth, I’m not sure “the credible threat of initiated aggression” being “considered “initiatory force” by NAP too” is “exactly the reason people would hire a PDA in an Ancap Society.” I mean, actual physical force is also a reason. Anyhow, the fact that demand for PDAs is motivated by the existence of threats does not mean the PDA market violates or fails to conform to NAP. The demand-side motivation (fear of threats) is irrelevant to the supply-side moral character (voluntary exchange and defensive force). To “conform to” the said principle, a market need only ensure that its transactions do not involve initiatory force.
Seventh, I would not go as far as to call insured events entirely unforeseeable. Analysis often involves quantifying the frequency of a loss event and its severity.
Eighth, saying occurrences “cannot be controlled by the company itself” needs specificity. It’s not as if they are just sitting there. For instance, property insurers require fire safety measures. The fact that the company “cannot have any control over earthquakes” is only an example of a weakness, not a defining one.
Ninth, you say that “what the anarcho-capitalist insurance/defense company approach offers is a fusion between insurance and defense.” This is only one of the various forms the company can take. Let’s see what David Friedman says in “The Machinery of Freedom:” “How might such agencies protect? That would be an economic decision, depending on the costs and effectiveness of different alternatives. They might limit themselves to passive defenses, installing elaborate locks and alarms, or they might take no preventive action at all but make great efforts to hunt down criminals guilty of crimes against their clients.” Aside, the insurance-defense fusion you describe is an extension of existing loss-prevention practices, not a radical departure from them. Also, you are overstating the degree of “control over the particular insurable event.” Do I really need to remind you that criminals are independent agents with their own motivations, information, and decision-making processes? Other than that, you are conflating beneficial control with harmful control when you say that “once this control is available, there is no doubt that it will be used in the best interest of that company.” The “control” that a defense company exercises over insured events (e.g. by deterring and preventing aggression) is precisely the kind of control that benefits clients. Consider, for instance, a health insurer that also provides preventive care: the insurer gains some “control” over health outcomes, and this control aligns with the client’s interest (fewer illnesses, lower costs). Tangentially, Hans-Hermann Hoppe writes in “The Private Production of Defense:” “for one, if an insurer could not prevent a crime, it would have to indemnify the victim. Thus, above all insurers would want to be effective in crime prevention. And if they still could not prevent it, they would want to be efficient in the detection, apprehension, and punishment of criminal offenders, because in finding and arresting an offender, the insurer could force the criminal—rather than the victim and its insurer—to pay for the damages and cost of indemnification.” It does not sound too bad to me! I mean, just remember Adam Smith when he says in “An Inquiry Into the Nature and Causes of the Wealth of Nations”: “It is not from the benevolence of the butcher, the brewer, or the baker, that we expect our dinner, but from their regard to their own interest. We address ourselves, not to their humanity but to their self-love, and never talk to them of our own necessities but of their advantages.” Plus, your “no doubt” claim is epistemically reckless because it ignores the market mechanisms that constrain corporate behavior. We are not talking about a monopoly. Let’s not forget that clients can switch PDAs. To quote Bruce L. Benson’s “The Enterprise of Law”: “In a private system, a seller of protection services has to convince customers that he is going to protect them before he even can begin amassing power. If he suddenly turns on his customers, they can hire another protection firm and the power dissipates.” Or incidentally: “But a security firm whose employees abuse the rights of citizens is not going to attract much business. An enforcer who fabricates offenses, intentionally prosecutes innocent people, practices entrapment, or waits for crimes to be committed is not going to receive many contracts once consumers of enforcement services catch on. After all, even though such actions may initially generate more work and profit for the enforcer, they also generate higher costs and poorer quality services for the purchasers of enforcement.” Also, consider Rothbard in “For a New Liberty:” “In the first place, any such favoritism will be highly unlikely, given the rewards and sanctions of the free market economy. The very life of the court, the very livelihood of a judge, will depend on his reputation for integrity, fair-mindedness, objectivity, and the quest for truth in every case. This is his ‘brand name.’ Should word of any venality leak out, he will immediately lose clients and the courts will no longer have customers; for even those clients who may be criminally inclined will scarcely sponsor a court whose decisions are no longer taken seriously by the rest of society, or who themselves may well be in jail for dishonest and fraudulent dealings.” Or this one: “The same analysis applies to the possibility of a private police force becoming outlaw, of using their coercive powers to exact tribute, set up a ‘protection racket’ to shake down their victims, etc. Of course, such a thing could happen. But, in contrast to present-day society, there would be immediate checks and balances available; there would be other police forces who could use their weapons to band together to put down the aggressors against their clientele. If the Metropolitan Police Force should become gangsters and exact tribute, then the rest of society could flock to the Prudential, Equitable, etc., police forces who could band together to put them down. And this contrasts vividly with the State.” Nevertheless, this problem of perverse incentive is worse in the context of the state. Providing both “insurance” and “defense,” the state exercises more control over the occurrence of threats than any private company would, with no competition whatsoever. Staying with Rothbard’s “For a New Liberty”: “It is particularly ironic that conservatives, at least in rhetoric supporters of a free-market economy, should be so complacent and even admiring of our vast military-industrial complex. There is no greater single distortion of the free market in present-day America.” And: “There is another reason why State aggression has been far more important than private, a reason apart from the greater organization and central mobilizing of resources that the rulers of the State can impose. The reason is the absence of any check upon State depredation, a check that does exist when we have to worry about muggers or the Mafia.” And: “Flynn predicted that America’s postwar policy would be ‘internationalist’ in the sense of being imperialist. Imperialism ‘is, of course, international . . . in the sense that war is international,’ and it will follow from the policy of militarism. ‘We will do what other countries have done; we will keep alive the fears of our people of the aggressive ambitions of other countries and we will ourselves embark upon imperialistic enterprises of our own.’ Imperialism will ensure for the United States the existence of perpetual ‘enemies,’ of waging what Charles A. Beard was later to call ‘perpetual war for perpetual peace.'” Also, I refer to Robert Higgs’ “Crisis and Leviathan:” “Relative to the economy, the government has grown enormously during the twentieth century. By conventional quantitative measures it is now three to six times as large as it was before World War I. Much of the growth has occurred during short intervals of national emergency, especially during the wars and the Great Depression. But even the growth that has occurred during normal times can often be traced to sources in preceding crises. National emergency can be seen to have been the fountainhead for the greater part of the growth of government particularly when the growth is viewed as expansion of the effective authority of government over economic decision-making. The ratchet applies not only to many aspects of governmental growth as measured by standard quantitative indexes; even more importantly it applies to the essence of the emergence of Big Government, the rise of government’s coercive power over economic life.” Finally, to quote Gustave de Molinari’s “The Production of Security:” “Now, if the producers of security are originally stronger than the consumers, won’t it be easy for the former to impose a monopoly on the latter? Everywhere, when societies originate, we see the strongest, most warlike races seizing the exclusive government of the society. Everywhere we see these races seizing a monopoly on security within certain more or less extensive boundaries, depending on their number and strength. And, this monopoly being, by its very nature, extraordinarily profitable, everywhere we see the races invested with the monopoly on security devoting themselves to bitter struggles, in order to add to the extent of their market, the number of their forced consumers, and hence the amount of their gains.”
Tenth, you seem to be forgetting about wealthy individuals living in remote areas when you say that “if there is no crime, people will not insure themselves against it.” The potential for crime also plays a role, not just experience.
Eleventh, I just want to correct your phrasing when you say that: “If there is no crime, people will not insure themselves against it, so no clients and consequently no revenues for the insurance companies. Note that if one has no clients, one cannot minimize costs in the mentioned way, because these costs exist only if there are clients.” I would suggest saying instead that firms lower prices when demand is weak—but when demand is zero, prices stop mattering and the market for insuring against a specific risk collapses.
Twelfth, the two opposing motives you describe apply to virtually every problem-solving industry. Doctors profit from illness, yet we do not conclude medicine is inherently self-defeating. Mechanics profit from breakdowns. Lawyers profit from disputes. Aside, why don’t we check what Friedman has to say about this in his book “The Machinery of Freedom”: “The main defensive weapon of organized crime is bribery. It works because policemen have no real stake in doing their job well and the taxpayers have no standard of comparison to tell them if they are getting their money’s worth. What is the cost to the chief of a police department of letting his men accept bribes to permit crime? In most cases, nothing. The higher crime rate might even persuade the voters to vote more money and higher salaries to the police department.” Whereas he says that “if employees of a rights enforcement agency accept such bribes, the situation is rather different. The worse the job the agency does, the lower the fee it can charge. If the customers of one agency find they lose, on average, ten dollars a year more to thieves than the customers of another, they will continue to do business with the inferior agency only if it is at least ten dollars a year cheaper. So every dollar stolen from the customer comes, indirectly, out of the revenue of the agency.” We can just say that the temptation to perpetuate crime is a diffuse, collective-level incentive; the reward for eliminating it is direct, immediate, and competitively decisive. Basically, the conflict of interest you identify is not solved by monopoly; it is maximized by it.
I would love to send you my feedback on your article. Would you mind?
Thank you for the detailed answers. The only problem is that they are more than I can practically answer. So, I will address some of the more important ones.
Please note that, for whatever reason, I do not receive notifications of your answers. And it’s okay for you to send me your comments privately.
Quote: ”The NAP condemns acts of initiatory force. It does not condemn markets that arise in response to the existence of initiatory force.”
You have a lot to explain and justify here! In particular, there are a lot of markets that exist only and exclusively due to the credible threat of initiated aggression, and the standard libertarian view is that these markets are not free. Examples: insurance markets imposed by government decrees, “green” markets (e.g., the CO2 quota system), etc. Lots of these exist, and they exist because if one does not do what the government says, one goes to jail (i.e., the credible threat of initiated violence is the driving factor). According to what you stated above, these markets are all OK. Do you or do you not support such markets?
Here is the problem with the above view: NAP does not specify where the initiatory force comes from! Government is just one source of initiatory force, and libertarians justly oppose such acts (which follow from NAP). But initiatory force need not come from the government! It can come from private individuals. And if we follow NAP, then this must be equally wrong. Thus, a market which exists due to private initiatory force must be non-free (non-NAP-conforming) by definition. And here is the problem: Under AnCap, people would get the services of PDA due to the justified fear of initiated force (being robbed, killed, etc.). So, again: How can you oppose the CO2 quota market and not oppose the PDA market? They are the same, just the “villain” is different.
Note that it does not matter where the initiatory force comes from in a particular market. It need not come from the direct market participants (e.g., sellers or buyers). It may be an external factor (e.g., the government decree, which forces buyers to engage with sellers in a way they would not wish to do). This, however, is exactly the case with the AnCap PDA too! Neither the PDAs nor their clients aggress against each other. It is the third (outside) factor, namely the presence of crime, that enforces the existence of such a market.
Quote: “Fourth, the conclusion that the contradiction exists is, hopefully, clearly wrong as we do not accept the broad definition of “caused by initiated aggression.” In fact, the latter would also render illegitimate any market for self-defense, locks, etc.”
Yes, the market for locks (house, bike, etc) is not free in my opinion. Note that (as I explain in my articles) this does not make it ethically wrong! And the reason is that without such a market, people would be worse off. Initiated aggression exists in the world, and until it exists, markets that are not free will exist. In short, I am not against the existence of such markets. I just state that they are not free according to NAP. It is your task to try to justify how such markets are NAP-conforming.