A Response (by David Friedman)

to a Rebuttal (by Jonathan Andreas)

of a Critique (by David Friedman)

of an faq (by Mike Huben)


When I wrote my critique of Mike Huben's anti-libertarian faq, it was my hope that Mike would write a rebuttal. I could then write a re-rebuttal, and ... . My ultimate objective was to use the web to produce a superior substitute for the threaded arguments common on Usenet newsgroups.

So far as I know, Mike has not yet responded. But Jonathan Andreas has. His response to my critique is available here. My re-response to many of his points is below. To make it a little clearer who said what, I have color coded everything:

Mike's original comments

My responses

Jonathan's rebuttals

My re-rebuttals.


Jonathan starts by explaining that:

I have linked the section numbers to MH's faq as DF did, but unfortunately DF did not put targets into the HTML of his critique, so I can only include a generic link to the top of his page.

You had only to ask; I have now added anchors to my critique.

1. MH: [Libertarians] are utopian because there has never yet been a libertarian society

DF: A utopia is an ideally perfect society, not merely a society that has never existed.

DF appears offended that people think libertarianism is utopian. However, if it is possible to describe any group as utopian, then that word applies to many libertarians.

It applies to many libertarians, many liberals, many communists, many syndicalists, even some conservatives. If Mike had written that some libertarians are utopian the statement would have been true but uninteresting.

MH's point remains that libertarians can not point to any example of a libertarian society (outside of dubious romantic versions of some pre-industrial societies like the tiny village society on Iceland).

It is useful, before criticizing ideas, to know at least a little about them. Iceland wasn't a village society--so far as we know, there was not a single village on the island. And my analysis of its legal institutions first appeared, not as "romantic versions," but as an article published in a peer reviewed academic journal.

DF's version of libertarianism (anarchy with property rights) is even farther removed from any historical example than more moderate libertarianism. If no libertarian society has ever evolved outside of fiction, then it is surely a utopian movement that has not even succeeded at creating a libertarian island or suburb or any community.

As I pointed out in my initial response to Mike, "utopian" doesn't mean "unrealistic" it means "ideally perfect." Consider the example I offered of modern mass franchise democracy. Was that a "utopian" idea in 1700, when no such society had ever existed?


2. MH: Are libertarians serving their own class interest only?

DF: What class interest? Libertarians are not a class in any economically relevant sense.

MH could easily clarify the class interest of libertarians in his faq. Define the economic class of libertarians as being those above the median expected total lifetime income. Certainly not all libertarians are well above the median expected lifetimeincome, but all the ones I have met are. It might be similar to say that about 90% of African Americans vote Democrat because they have some class interest. Obviously not all Democrats are African American and not all wealthy people are Libertarian. The vast majority of wealthy people see the benefits they have received from the economic system created in part by our government and feel that they have some duty to pay some taxes. If the super-rich did not feel the duty, they would have the resources to be able to dodge much more tax than they do. Some wealthy people feel no such duty and there are many examples of these elites (Marc Rich for one) who go to great lengths to pay very little tax. As Leona Helmsey said, according to her world view, "Only the little people pay tax."

Your heroic efforts to defend Mike even when his position is indefensible demonstrate courage--but not sense. You have just asserted that his reference to libertarians serving their class interest can be defended by:

A. Defining a class that includes half the population and has none of the characteristics usually associated with social classes.

B. Noting that you think a considerable majority of libertarians are in that class.

C. Noting that you think the "vast majority" of people in that class (actually, in a subclass to which your arguments ought to apply even more strongly--"wealthy people") don't perceive their interest as what you claim Mike claims is the class interest that motivates libertarians.

In the process you have implied, as Mike probably intended to imply, that libertarians are to be identified with elites, super rich, the wealthy, etc. You have, in other words, first defined your class broadly, so as to be able to claim that most libertarians are in it, and then switched to a much narrower (and more easily demonized) subgroup.

Most libertarians applaud and encourage self-interest. Naturally, wealthy people who expect to pay above the median lifetime taxes would find it in their short-term self-interest to abolish them and find Libertarianism an attractive ideology to justify dodging taxes.


If, as you and Mike believe, taxes produce benefits much larger than their costs, then people who expect to pay above the median would still want to keep taxes. If, as I believe, taxes produce benefits much smaller than their costs, then even people who expect to pay below the median would want to abolish them. And if, as most defenders of graduated taxes claim, people with higher incomes get more benefit from government activity than people with lower incomes, then even in a society where each tax dollar by some odd coincidence produced exactly a dollar's worth of benefit the people paying higher taxes wouldn't consistently support abolishing them.

You, like Mike, are trying to imply that libertarians are evil people who make dishonest arguments for selfish reasons--but you are less ashamed of making that claim than he is and so make it more explicitly.

So far as dodging taxes, that is a profitable activity for those who can get away with it whether or not they are above the median.


3. DF: Mike seems to forget in this passage that he himself is, by his definition, an evangelist.

Perhaps MH should expand his definition of Evangelist from "(those trying to persuade others to adopt their beliefs)" to *only* include people who as he later states "tend to be more interested in effect than in accuracy." Then DF's critique of this point would loose nearly all its weight.

As would Mike's original argument.

You are confusing your beliefs about other people's arguments with theirs. If I evaluate your arguments by whether I think they are accurate, you fit your definition of an evangelist, since most of them aren't. But I have no doubt that you think they are accurate. Similarly, I expect that even libertarians who you and Mike think are making bad arguments--even libertarians who I think are making bad arguments--for the most part believe that they are making good arguments, hence do not fit your definition of evangelists.

As you would know if you were more familiar with libertarianism, a small number of libertarians have for years been trying to persuade other libertarians to select those (correct) arguments that will persuade other people in preference to those that will offend them--with very limited success.

5. MH is correct that the Government is the foremost defender of our freedoms and rights. DF also correctly points out that it is also a foremost infringer of rights. Does Government create and defend more freedoms than it destroys? It is hard to tell.

Hence MH's argument, which implies that it is easy to tell, is wrong.

Part of the problem is defining what rights are. Rights are social constructs.

If you think that is all they are, let me suggest two things worth thinking about.

A. If a society decides that torturing dissenters to death is legal, do they have a right to do so?

B. The analysis offered in my "A Positive Account of Property Rights."


DF has a good point that governments kill a lot of people, but it is unfortunate that DF does not give any data to back up his claim that governments kill more citizens than private murders.

Why should I do all the work?

You can find the data on government killing in the 20th century in Rudolf J. Rummel, Statistics of Democide: Genocide and Mass Murder since 1900. I don't have a similarly complete source for private murder, but one can make at least rough estimates.

The U.S. murder rate peaked at about 10/100,000/year, generally regarded as a very high level by both historical and international comparisons. Assume the world average over the 20th century to be a tenth of that. Assume the population of the world averaged three billion people over the 20th century. That gives you an estimate of three million private murders over the century--roughly one fiftieth of Rummel's estimate for government murders. A factor of fifty allows lots of room for varying the detailed assumptions.

Do democracies kill more citizens than private murders? Probably not. It is important to distinguish between the government of authoritarian communist China and that of peaceful democratic Costa Rica. ...

Germany was a democracy for most of the fifty years preceding Hitler's takeover. Indeed, Germany under Bismark was widely regarded by progressives as a model to be imitated, and a number of important institutions in modern societies are modeled on it. Cambodia was a reasonably peaceful (although not democratic) society prior to the events that led to the Khmer Rouge killings. You can't judge whether governments, on net, do more good than damage, or violate more rights than they protect, by only looking at the good governments in the good years.


DF is disingenuous to compare the crime rates of England in the 1700 to that of today and say that it worked better then because of private law enforcement.

I might have been disingenuous to have said that, but I didn't. What I actually wrote was:

"There is also some evidence that the murder rate in 18th century England, where most of what we think of as law enforcement (catching and convicting criminals) was private, was not radically different from the rate a century later, under modern institutions."

I count three false statements explicit or implicit in Jonathan's one sentence summary of that.

"not radically different" is not equivalent to "worked better"

"a century later (than the) 18th century"=19th century not "today"

"There is also some evidence that X" is a much weaker claim than "say that X."

Jonathan seems to have difficulty with such fine distinctions.

and institutions that worked in hyper-religious pre-industrial village society probably would not work in today's gun infested urbanized society.

England gots its first police force (in anything close to the modern sense) in the 1830's. During the preceding century it was not hyper-religious--that was the century when Hume, arguably an atheist, could defend established religion on the grounds that it "bribed the indolence of the clergy." The right to bear arms was well established in English law by the end of the seventeenth century.

DF also gives the example of Saga period Iceland (going back 1000 years or more) as a model for government and law enforcement today. This is ludicrous.

It might have been if I had, but I didn't. Mike's claim, which I was answering, was that "The foremost defenders of our freedoms and rights, which libertarians prefer you overlook, are our governments." My response was that:

This is presented as if it is a fact when it is actually an interpretation, and a highly contestable one at that.

Mike's interpretation of the evidence implicitly assumed that rights could not be protected save by government. I was providing a counterexample--one example of a society without either government law enforcement or horrendously high crime rates.

If you want my model for today, you will find it in part III of The Machinery of Freedom. That was written before I knew anything about Icelandic legal institutions. But you might describe it as an updated version of those institutions, just as you might describe a modern market economy as an updated version of a village market.

There are more recent (and close to home) examples of societies in the history of the American West that had no government expenditure on law enforcement, with high levels of rights and property rights protection. However, the vigilante justice of the period did not follow due process and was extremely harsh (immediate death upon apprehension for alleged theft). Few people would trade our expensive modern criminal justice system for the fiscally cheap vigilante justice systems of the pre-government American west.

You might want to look at Anderson, T. and Hill, P. J. "An American Experiment in Anarcho-Capitalism: The Not So Wild, Wild West," Journal of Libertarian Studies. Vol. 3, No. 1. 1979 for a good antidote to the popular moviegoers' view of that period.

6. DF would have us believe that public defenders, the Constitution and the Bill Of Rights are "almost entirely government efforts to protect rights and freedoms from infringement by the government." It is hard to imagine that DF or any reasonable person believes that government is "almost entirely" the only thing that infringes upon our rights.

It is hard to imagine that you, or any reasonable person, could read what I wrote so carelessly. "public defenders, the Constitution and the Bill of Rights" are not the only things that could protect rights. Hence observing that those things exist mainly to protect against infringement by the government does not imply, or even suggest, that only government can infringe rights.

How does DF measure the percentage that these legal institutions defend against the government and how much they defend against private parties? Only a very small percentage of legal cases claim the government as defendant.

But all criminal cases claim the government as plaintiff--that is part of the definition of a criminal case. Public defenders are attorneys provided by the government to defend people who are accused of crimes and can't afford a lawyer. Hence they exist entirely to defend people against prosecution by the government.

It is also obvious to anyone that this list of government institutions defends against all infringements by anyone including the government, so this point of DF does not refute anything in MH's faq.

The Bill of Rights consists primarily of descriptions of things that Congress may not do. Public defenders exist entirely to defend criminal defendants against government prosecutors. It's true that the the Constitution exists in part to prevent rights violations by foreign governments, but mostly it sets up and constrains the structure of the federal government.

Insofar as government, under the original Constitutional scheme, was involved in protecting rights against non-governmental infringement, that was the job of state and local governments--which were not established by the Constitution or the Bill of Rights.

8. DF goes to the heights of absurdity to claim that libertarians achieved the "abolition of slavery, the institution of large scale free trade, the destruction of guild restrictions on employment--most of the progress of the 19th century". Statists (assuming that there is someone who is willing to go by that libertarian umbrella label) could more easily assert the same claim for themselves. VERY few of the many people who worked to end slavery (such as Abraham Lincoln) were anywhere near the political stripe of DF and all of these examples of progress were the direct result of government action rather than some kind of market or libertarian force. DF might as well also claim that the spread of democracy, extension of life span, eradication of smallpox, and the huge improvement in the standard of living of the past century are also "progress" and are therefore libertarian accomplishments.

What I wrote was that

"libertarianism, in its earlier and somewhat more moderate incarnation as classical liberalism, has a historical track record ..."

Are you denying that the changes I described were the work of classical liberals? Are you denying that classical liberalism was an earlier and more moderate version of libertarianism?

It's true that the abolition of the slave trade depended on government power (the British navy)--motivated in part by classical liberal arguments. But my other two examples--the shift to free trade and the destruction of guild restrictions on employment--were both cases of a government that had been doing things (taxing imports, enforcing guild monopolies) ceasing to do them. I am puzzled as to how you can describe that as "the direct result of government action." If the U.S. government were to end the war on drugs, would you describe that too as government action?

Have you ever heard of the Corn Laws? The anti-Corn Law League? Cobden? Bright?

I have, however, changed "the abolition of slavery" to "the abolition of the slave trade" in my reply to Huben, since that is what I was actually referring to.

9. DF suggests "that we reduce government expenditure to the level that can be supported by taxing [income from unproduced resources]." which he claims are only a few percent of national income.


It is easy to save a few percent of the national budget by cutting popular libertarian enemies like NPR funding, but the big expenses are often things that most libertarians would secretly like to keep such as defense.

As it happens, I would be happy to drastically cut defense spending. But in any case, as you can easily check from the Statistical Abstract, defense spending is only a little over ten percent of total government spending (state, local, and federal combined). Education is about fifteen percent. Public welfare expenditure (not including social security) is roughly comparable to defense expenditure.

Ironically, DF says he would accept (as a compromise) a tax on unproduced resources like land. Until now he has been arguing that taxation equals theft and that the government does not have the right to tax.

More precisely, I have been rebutting Mike's arguments against people who make that argument.

Now he is willing to accept billions of dollars of property tax. What happened to the righteous indignation against government and taxation? If DF really believes taxation is the moral equivalent of violent theft via men with guns, then no amount would be OK. How is it possible for a libertarian to morally justify accepting this theft but not others?

I haven't morally justified any taxation at all. I merely proposed to Mike a compromise between what he thought was justified and what I thought was justified. In case it wasn't obvious, the point of the proposal was that if we limited government to what the argument he had just made justified, we would have to eliminate most of it--a "compromise" biased heavily in my direction.

It is also ironic that DF chooses to live in a city and a state with relatively high property tax, sales tax, income tax and extensive local government services and regulations. If he were really convinced that taxation is the moral equivalent of violent theft via men with guns, then why doesn't he take some very simple precautionary steps to avoid it and move to a city and state with less taxes and regulation? Most people go to great lengths to avoid violent crime. Perhaps he doesn't really believe his own analogy.

You could choose to live in a country with a lower murder rate. Does it follow that you don't really believe that murder is immoral?

10. DF is correct that the origin of the "social contract" is problematic. However, it is a convenient construct to explain and give moral justification for how society has evolved even if it is just a metaphor. However, it is no more problematic than property rights or laws. Laws are also problematic without some kind of social contract. Why should I obey any laws? I did not sign any social contract.

Do you believe that there is nothing wrong with murdering people--assuming you have not previously promised not to? I didn't think so.

More generally, the argument you are offering is internally inconsistent. In order for a social contract to be morally binding, you need to assume that before signing it we alreadly have a moral obligation to keep our contracts. Why are you willing to assume that pre-existing moral obligation but not the moral obligation not to murder people?

11. 12. 13. Again, a social contract is an arbitrary social construct, but the same thing is true of property rights or laws.

All laws? You believe that laws against rape and murder are merely "arbitrary social constructs?" If so, then if the law were the other way murdering and raping people would be just fine. Is that your view?

Libertarians just get more fixated on the taxation clause in the social contract than the property rights clause. If the government creates the infrastructure that underpins our economy such as money, the legal system and property rights, does it not have property rights over the use of that infrastructure? Why can't it then tax the use of these things as it sees fit?

I have no objection to the government taxing the use of government money--providing that it lets us use other money if we prefer. I don't even object to the government charging fees for using its courts--as long as it leaves us free to use other courts if we prefer.

If you copy CDs and freely distribute them at your own expense to the poor (who probably would not buy them anyway) MEN WITH GUNS will initiate force and put you in jail. Why do libertarians get more excited about tax law than copyright or other laws?

That depends on the particular libertarian; some of them don't like intellectual property laws either.

Why is it that, when your argument is about "other laws," you conveniently choose intellectual property laws for your example--a category of law that many people do not find morally persuasive. You could have made your argument just as well by starting out "if you go around murdering men, raping women, and torturing children to death, MEN WITH GUNS will ... ." Why didn't you? You might want to think about the question, in order to decide whether you really believe the arguments you are making.

14. MH: Some libertarians make a big deal about needing to actually sign a contract. Take them to a restaurant and see if they think it ethical to walk out without paying because they didn't sign anything.

DF: The act by which one agrees to an implicit contract is an act that the other party has the right to control

 DF's assertion makes no sense. In any contract, both parties must agree. Thus with an implicit contract, DF is arguing that both parties must have the right to control the act. In the case of the restaurant, who has the right to control whether food is exchanged for money?

Both parties have the right to veto the act, neither has the right to insist on it--as with any other voluntary transaction. If you don't want to sell to me, you don't; if I don't want to buy from you, I don't. Why do you find that puzzling? It is the basic rule by which we conduct a large part of our lives.

Consider a different voluntary transaction--marriage. Do you have any problem understanding in what sense both parties have a right to control the act?

One has the right to control whether they stay in a country or not.

I have the right to control whether I stay in a country or not--hence I have the right to say to the government (or anyone else) "I will only stay in the country if you meet the following conditions"--and then emigrate if my conditions are not met. But Mike's argument requires that the government have the right to decide whether I stay in the country or not.

Note that I wrote "an act that the other party has the right to control." If my act is staying in the country, the "other party" is the government, not me.

15. [some legitimate points by Jonathan about problems with the origin of property rights omitted--the response would have to be way too long for this context].

The important question for libertarians to answer is how do they provide a rule of law without government or government territory? Without some sort of territorial rights, there is no way to have laws. Does DF want to create a society of men with guns in which everyone must enforce their own individual contracts with guns?

I answered that question in a book published in 1973 and still available from Amazon.com. You can even find a few chapters on my web site.

For a brief hint ... . Do you believe in government run agriculture? If, as I assume, the answer is no, does that mean that you want to create a society where everyone has to grow his own food?

So far as the claim that without "territorial rights" there is no way to have law, that's absurd. The Law Merchant arose without territorial rights, as did the law of the sea. There are lots of historical examples of societies where what law applies to a dispute is determined, at least in part, by something other than what territory the dispute occured in--including ours.

There are also many examples of government rights evolving without conquest. For example, the powers of the federal government of the US, and the EU evolved in large part without conquest. The transferal of power from the British monarchy to their democratic government of today also evolved largely without conquest.

Those are both cases where one government transferred rights to another, not where individuals gave rights to a government. And in any case, the big shift of power to the U.S. federal government was the result of a long and bloody war, as was the big shift of power from the English monarchy to parliament.

16. ... 

When democratic governments take away taxes, they do not privately consume them like criminals would (increasing criminal consumption is not necessarily a bad thing, but most people are morally opposed to it). They primarily redistribute the resources to other individuals and produce (mostly) public goods for the benefit of all citizens. By equating criminal theft and taxes, DF is wrongly assuming that criminals have an incentive to be benevolent like Robin Hood and steal for the benefit of the public.

You have it backwards. It isn't that I assume criminals are benevolent but that I don't assume governments are.

Your claim that governments redistribute is true--but so do criminals. One common outcome on the political market is that interest groups use government force to redistribute to themselves--just as criminals use private force to redistribute to themselves.

Your second claim is that governments produce mostly public goods for the benefit of all citizens. You asked me to provide support for the claim that governments in the past century killed more people than private individuals--a statement that is true with a safety margin of well over an order of magnitude. Would you like to provide support for your confident claim about what government produces and who benefits?

I would have said that many goods governments produce are private goods and that, with the possible exception of national defense, most government activity does not benefit all citizens, or even most citizens. Governments quite routinely do things--the farm program and all tariffs come immediately to mind--that injure most citizens. They do them in order to buy the political support of particular interest groups.

17. MH: (1) If taxes are eliminated, you'll need to purchase services that were formerly provided by government. (2) If taxes are eliminated, the economics of wages have changed, and wages will change as well.

 DF's response to MH's point (1) is that, "most of the services provided by government cost far more than they would if provided privately". However, the private sector will not provide most government services at all. Most government services are public goods and it is very difficult for the private sector to make a profit from providing public goods.

Both halves of the statement are false. Schooling is the largest single expenditure of U.S. governments--and it is a private good in the ordinary sense of the word. The private sector quite routinely makes a profit providing public goods--consider radio and television broadcasts, which unlike schooling really are pure public goods.

Many economists argue that when the economy (government or market) does not provide enough public goods, the efficiency of the economy declines and everyone is poorer.

If public goods are produced at suboptimal quantity or quality, people are on net poorer than they would be if public goods were produced in optimal quantity and quality, all else being equal. But to get from that to "therefore government should produce all goods that the market underproduces" you need at least one further step--an argument to show that the political marketplace will actually yield production at optimal quantity and quality. No such argument exists.

DF also gives three examples of arguably bad regulation to demonstratehow bad government is. Unfortunately, whenever there are laws created by humans, some of it will be bad regulation. A libertarian government will not change that. American citizens will always have to be vigilant to put pressure on lawmakers to change laws regardless of how big or small the government is.

1. My libertarian society will change that, because it won't have a government to make bad regulations. I agree that a libertarian minarchy will have to worry about bad laws for the same reason our current system does. That's one of the reasons I am an anarchist.

2. Perhaps we should also solve our energy problems by being vigilant to make sure water runs uphill, thus allowing us to power our cities from generators located on circular rivers.

Or in other words, I am offering a theory of what government will do, you are offering pious hopes inconsistent with what we know of human behavior.

18. 19. DF is correct that the government was a smaller share of the economy in the 19th century. However, most Americans demand more expensive service from their government today and it is no surprise. People are several times wealthier now and our demand and consumption of all goods (including those produced by government) has also increased several times.

What you just agreed to is that the government share of the economy increased. If government spending merely increased in proportion to private spending, the government share would remain constant.

Many of the services that government provided in the 19th century, such as defense, education, and public health, have become much more expensive relative to other sectors of the economy such as manufactured goods and commodities which are generally cheaper now (in real terms) than in the 19th century.

Why is it that government activities become selectively more expensive to produce--by a factor of nearly four, if your argument is to explain the growth in government? How come none of these services exhibit economies of scale? Wouldn't you think that, as they became relatively more expensive, we would substitute away from them? Don't you detect a faint whiff of special pleading?

20. MH: "Self government" is libertarian newspeak for "everybody ought to be able to live as if they are the only human in the universe, if only they believe in the power of libertarianism."

DF: You don't need the power of libertarianism--standard neoclassical economics gets you most of it. To first approximation, the price system allows each individual to use his resources to achieve his objectives without imposing net costs on others--for details see the chapter on " What is Efficient" in my (online) Price Theory.

DF has a lot of great ideas for economics, but in this case, DF's economics is not so much "standard neoclassical economics" as it is neo-economics. Neoclassical economics does not lead to libertarianism in the eyes of the vast majority of economists. This is obvious because the vast majority of economists are not libertarians.

The vast majority of American economists are more libertarian on a wide range of issues than American academics in general, as was revealed by a couple of AER articles based on polls a good many years back. I agree, however, that very few American economists are as libertarian as I am.

But if you actually read what I said, instead of inventing claims to refute, you might notice that I didn't say or imply that most economists were libertarians. I merely said that standard neoclassical economics solved one of the problems that Mike raised with libertarianism--the problem of how individuals can live "as if they are the only human in the universe" when they are actually part of a complicated interdependent society. Explaining how it does so requires more than a paragraph or two, which is why I pointed the reader to a chapter in a webbed book.

I checked DF's link to his textbook chapter on efficiency and found it conveniently omits several preconditions to efficiency which other, less ideological, textbooks do not. Not everyone are price takers, there are transaction costs, and there are externalities. ...

"Throughout the argument, I have assumed that everyone concerned--firms, owners of the factors of production, and consumers--is a price taker. If even a single participant in the market is not, then somewhere in the chain of argument a link fails and we can no longer prove efficiency. " (Price Theory, Chapter 16)

If you are going to make statements about books, even chapters, it is useful to first read them.

You will find transaction costs and externalities in Chapter 18.

Another major flaw with using economic efficiency as a moral yardstick is that it says nothing about justice. ...

If you are criticizing my response to Mike Huben, this is irrelevant--the passage you are responding to says nothing at all about using economic efficiency as a moral yardstick. If you are criticizing my book, you ought first to read it; the issues you are raising here are discussed in Chapter 15.


25. DF: A large fraction of the arguments for government regulation of individual action depend on the implicit assumption that individuals act on their own self-interest under conditions of limited information in market contexts, but that government actors are fully informed and benevolent--with no theory to derive the latter from the former.

There is no more need to assume that government is fully informed and benevolent to derive the benefits of government than there is need to assume that individuals are fully informed and benevolent to derive the benefits of markets. A very simple theory from late economist Mancur Olson to derive the beneficial impact of government is as follows.

(omitted argument demonstrates that governments will be less destructive than roving bandit gangs, since they have more secure property rights, which is correct.).

Later some nations evolved into democracies. These can be seen, in the least charitable light, as dictatorships of the majority which try to rob the minority. However, the majority (working in self-interest with limited information) will now decide to provide even more public goods in order to increase the income of the majority. ...

You will find a fairly detailed explanation of some of the reasons why this doesn't work in Chapter 19 of Price Theory. I expect Olson would agree with most of it, but unfortunately we can no longer ask him.

Furthermore, the roughly 190 national governments of our world compete with each other for productive citizens and for military and economic power. This competition provides some check on the excesses of governments. Factions within countries also compete for power. Generally, the more broadly distributed the power is in a country, the better the government will be at providing public goods as if it were purely benevolent.

Chapter 19 provides a rather more detailed analysis, mostly borrowed from Becker, of the result of factions within countries competing for power. I certainly don't deny that there are some checks on the excesses of government--if there weren't we would be a lot nearer starvation than we are. But you are jumping from the claim that government isn't infinitely evil, which is true, to the conclusion that government will predictably tend to exercise its power in a fashion that on net increases economic efficiency. And the comparision here is not "government versus bandit gangs" but "government vs laissez-faire"--otherwise you (and Mike) would be arguing for libertarian minarchy, and you aren't.

...The incentive for the ruler of the government is self-interest to maximize long term revenue by maximizing economic growth.

That might work for an immortal absolute ruler, assuming he was competent, but that isn't the system we are describing. In our system, political actors have insecure property rights in their political power, leading to precisely the problem you earlier described in the context of bandits. And there isn't "a ruler," resulting in some of the other problems that I describe in chapter 19.

Besides, even if the rest of your argument were true, you are confusing maximizing tax revenue with maximizing human welfare. Suppose there is some public good which benefits people not by increasing their future output but by increasing their present happiness. Happiness isn't taxable, so why should your hypothetical ruler bother to spend a penny producing it? Wouldn't his optimal policy be to impose the revenue maximizing tax rate on it?

27. DF: There are real examples of more or less libertarian societies, and of societies that in particular respects were entirely libertarian, so we do have real world evidence to go on.

Even if you accept DF's nostalgic version of history, DF doesn't give any "real world" examples that are more recent than the 19th century.

Hong Kong is quite a bit more recent. The U.S. has mostly private radio and television broadcasting; the U.K. for a long time had only public. The U.S. had private monopoly phone systems; European countries had government phone systems.

Or in other words, you ought to be be able to provide for yourself lots of modern examples of societies that differed in how libertarian they were in one dimension or another.

As to my "nostalgic view of history," I believe that all of my historical articles are available on my web page. Perhaps you could point out the "nostalgic" bits.

As I said before, we libertarian-skeptics are waiting for some brave libertarians to put their money where their mouth is and start creating a small libertarian community. ... Then if that works, they could buy (and subdivide) a large ranch (or small island) that is distant from government tentacles and begin a community of libertarians living in peace and harmony with markets and property rights, but no local government and no local taxes.

There have been a number of attempts along those lines. The problem is that existing governments are very reluctant to sell property along with sovereignty--and if they do, it is difficult to prevent them from reneging on the deal. Hence nudging existing societies towards being more libertarian seems, so far, a more promising approach.

People like the unabomber or Timothy McVeigh also espouse similar rhetoric as their philosophical underpinning.

I can't speak to McVeigh, since I don't think I have read anything he wrote, but the Unabomber's rhetoric was anti-technology, not libertarian.

Much popular libertarian rhetoric is a subtle revolutionary call to arms.

Much liberal rhetoric is, in the same sense, a subtle justification for poor people mugging rich people. In both cases, the fact that the rhetoric can be used that way tells us very little about whether the underlying arguments are true or false.