The Non-Non-Libertarian FAQ

The Non-Non-Libertarian FAQ

aka The webpage you will always remember for changing your political views, maybe 🙂


0. Introduction

A. Economic Issues

1. Externalities
2. Coordination Problems
3. Irrational Choices
4. Lack of Information

B. Social Issues

5. Just Desserts and Social Mobility
6. Taxation

C. Political Issues

7. Competence of Government
8. Health Care
9. Prison Privatization
10. Gun Control
11. Education

D. Moral Issues

12. Moral Systems
13. Rights and Heuristics

E. Practical Issues

14. Slippery Slopes
15. Strategic Activism
16. Miscellaneous and Meta




0.1: Who are you, what is this?

For me, see in the About section of this site.

This is a response to Scott Alexander’s Non-Libertarian FAQ, initially written in 2010 and revised in 2013. Note that his FAQ is not called the Anti-Libertarian FAQ. In the same spirit, this FAQ does not defend libertarianism, but criticises arguments offered against libertarianism.

0.2: Are you a libertarian?

That depends on your definition of libertarian. I’ve never been really comfortable with any ideological X’s in ‘I am an X’. But you can get an idea of what I think by reading Michael Huemer, Loren Lomasky, Robert Nozick, Matt Zwolinski and Jason Brennan.

On statism, I define statism as the belief that the State has political authority, or that the law ought to be obeyed just because it is law, whatever its content. Political philosophers nowadays reject this, so statists are on the wrong side of philosophical progress. However, arguments still remain for State commands being right for other goals. That is, laws against murder are nonproblematic, but that’s because murder is wrong, and the State enforces its ban. Also, there may be other arguments for the State that do not rely on political authority, namely consequentialist arguments.

Then, most non-libertarians (and some libertarians, curiously) are statists as defined above. Scott does not believe in political authority: he is a consequentialist. Hence, he is not a statist. He would support the state only if it is the social-institutional arrangement most conducive to the Greater Good™.

0.3 Do you hate non-libertarians?

Hating is irrational, and doubly so in intellectual inquiry. The mental state of ‘hating’ not only makes you feel worse, but it also induces you to think in biased ways regarding that which you hate.

Now, Scott says that he likes libertarianism as a set of hands-off policies that can be argued to work better than the alternatives, and that he dislikes the libertarianism that “rather than analyzing specific policies and often deciding a more laissez-faire approach is best, starts with the tenet that government can do no right and private industry can do no wrong and uses this faith in place of more careful analysis.”.

He addresses the FAQ to the second type of libertarians. I would be a third type: I do think libertarianism works better in general (based on empirical and theoretical evidence)  and that the situations where it doesn’t work as good do not justify the use of force to correct them. If there were situations where it worked really poorly, then coercion would be justified. That is, I believe States can do many things right, and that the private sector can do many wrong things, but that by itself doesn’t say anything about whether there should be a State, or if there has to be one, what should it be doing.

One thing is economics and history, and the other is moral philosophy. In this FAQ we will deal mainly with the former, as Scott has another FAQ that deals with the latter: the Consequentialism FAQ. Although I’m not a consequentialist, I think consequences matter greatly. This FAQ mainly deals with empirical issues, so it isn’t relevant for my purposes what the correct ethical theory is. To add to the prima facie unimportance of this debate at this point, many libertarians are consequentialists or utilitarians, so if consequentialism is true, it may end up leading to libertarianism anyways.

0.4 Will this FAQ prove that the free market is always better than the government?


What I will argue is that free markets are very good. Even better than most economists think they are. And in the long term, coupled with non-market institutions and a proper definition of property rights (to solve externalities), this is a (meta)system that is very hard to criticise by comparison to an alternative system.

0.5 Why write a Non-Non-Libertarian FAQ?

Scott Alexander has written a document for which there is no publicly available (high quality) critique, and given that it is well argued, and has a friendly tone, some have taken the document itself, and the absence of critique for three years as proof that libertarianism is wrong.

If a person as honest, rational and rigorous as Scott has rejected libertarianism in a non-strawmanning way, and no one has replied to his arguments, doesn’t it count as a strong prima facie case against libertarianism? It does! There is one such reply (not very good, imho), and David Friedman emailed Scott another, also available here. But still there is no publicly available good response to the entire FAQ, point by point. Of course, there are good argumentations for libertarianism elsewhere (see here for philosophy, see here for practical issues), but those are not specific to this FAQ.

That said, Scott Alexander admitted some years ago that he wasn’t sure he’s a non-libertarian anymore, and that the FAQ is aimed at non-smart libertarians, so writing my critique may seem, and I think it will seem, like an overkill.

Finally, I cannot let well argued arguments to stand unanswered if I disagree with them. If one wants to keep holding beliefs in disagreement with his FAQ, one must produce reasons for it. I make my list of such reasons available to everyone to move  the debate forward.

0.6 How is this FAQ structured?

The same way as Scott’s. I recommend you to read mine in parallel with his, as the questions in my FAQ are summaries of Scott’s answers, more or less. My want could even make little sense without this parallel reading.

As an additional comment, in this paragraph, Scott alludes to the tendency of libertarians to resort to economic and moral arguments. If morality dictates libertarianism, why bother with consequences? If libertarianism is the system that works best, period, why bother saying that things are morally wrong?

This critique was initially advanced by Jeffrey Friedman (1997), and was debated afterwards (Palmer 1998, Friedman 1998). Recently, Zwolinski and Tesón made this explicit:

While an appeal to self-ownership is probably the most famous libertarian strategy for defending rights of private property, it is by no means the only one. Some libertarians have sought to base their defense on different but equally deontological grounds such as an appeal to negative liberty, or to autonomy. Other libertarians have sought to  provide a more consequentialist justification, by showing how private property replaces the zero- (or negative-) sum transactions of the commons with the positive-sum  transactions of a market economy.  Or by showing how rights of private property are instrumentally valuable in securing other important civil and political rights. For the most part, though, individuals in the libertarian intellectual tradition have refrained from drawing sharp distinctions between consequentialist and deontological arguments, believing instead that respect for private property is both a moral duty and a wise social policy.  Libertarians, especially those outside the discipline of academic philosophy, have thus tended to help themselves generously to both sorts of arguments, without always distinguishing clearly between them. […]

That fact that justice and utility both point in the same direction in so many libertarian arguments is surely a fact that calls out for explanation. One possible explanation is the skeptical one. The reason justice and utility line up for libertarians, this line of reasoning goes, is because libertarians frame their arguments to fit their conclusions, rather than the other way around. See, for a somewhat sympathetic expression of this skepticism, Jeffrey Friedman, “What’s Wrong with Libertarianism?,” Critical Review 11, no. 3 (1997).
However there are other, less skeptical explanations available as well. For instance, perhaps justice and utility do not come into conflict because the content of justice is partly a function of utilitarian considerations? See, for a discussion of many such possible explanations, and an endorsement of one, Roderick Long, “Why Does Justice Have Good Consequences?,” [Ed. See also this and this]

I don’t have a defined view on this matter yet, but I will say that if libertarianism had very bad consequences, it ought to be abandoned. Also, if libertarianism had very good consequences, but it were highly unethical, it ought to be abandoned as well.

Part A: Economic Issues

The Argument

Every transaction affects not just the seller and buyer, but everyone else, and that it assumes that consumers have coherent wants and demands. Markets have problems: externalities, coordination problems, irrational choice, and lack of information. Given this, a regulated market trumps a free market.

The Counterargument

Yes, voluntary trade makes people ex-ante better off necessarily, but this does not presuppose all the conditions of classical perfect competition. Perfect competition gives sufficient, not necessary, conditions for efficiency.

It is true that trade occurs not in isolation. Precisely this will be the key to answer the challenges the FAQ raises. Also, trade doesn’t need to be between two people, you can have multilateral trade. As long as everything is voluntary and property rights are well defined, the free market delivers, it will be argued.

1. Externalities

1.1 What is an externality?

Externalities are costs imposed upon people who were not participants in the transaction that originated said cost. While some economists like to derive justifications for government interventions from externalities, we should remember that there are quite a few externalities no one advocates for fixing, and yet some are as externalities as pollution.

One is the psychological distress non-whites cause to white supremacists. If that externality has to be fixed, non-whites would be taxed for existing, and the white supremacists would be compensated, or something like that. Instead, most of us agree, and rightly so, that white supremacism is wrong, and that what ought to happen is for them to change their preferences. This also solves the externality by making it vanish.

Another one is procreation: bringing more children to the world alters the distribution of traits (intelligence, health, creativity, kindness, etc) in society, and that affects everyone (Anomaly, 2014). To fix it, we may require positive (unproblematic) or negative (problematic) eugenics, or family planning by some authority.

Most people would not endorse taxing non-whites to compensate white supremacists, or curtailing reproductive rights in such a way as to ensure every child is not a walking negative externality.

Also, we wouldn’t want to fix pecuniary externalities: externalities that affect others via the price system. If I buy orange juice, I marginally raise the price of orange juice for the rest of consumers, so it affects them negatively. But that loss for them is balanced by an equivalent gain (in money) for orange juice producers.

What (negative) externalities are usually the ones that most people would like to fix? Those that are basically violations of rights, and that is required by libertarianism.

There are also some positive externalities, but arguments that begin from there to justify coercion to fix them are harder to make.

1.2 But the libertarian ways to solve externalities fail, unless you propose something that basically is a government

The externality described in the FAQ would be solved by a court. You would kindly ask the wasp farmer to stop having his business there because he’s annoying, and if he doesn’t, you sue him. Alternatively, there could be private organisations that could to the suing for you. The court would assess if the nuisance caused by the wasp farmer is a rights violation or not. It probably would: people in the neighbourhood were there before, and the presence of the wasp farmer leaves them worse off in a very noticeable way.

Alternatively, another institution to fix that is a Homeowner Association (HOA). These are not governments, but provide most of the functions that you would like local governments for.

Another one is covenants, binding agreements that are transferred with the property of the house. Why would builders include prohibitions against wasp farming in the buildings in their newly built neighborhoods? Because people would value such neighborhoods more, obviously. This is also a foundation for libertarian zoning restrictions. For more info on this, see Beito, Gordon & Tabarrok (2009)‘s excellent book on the topic. These may not be used nowadays, because the government is already in the business of doing that, but in the past, it wasn’t that rare.

1.3 People won’t boycott companies that pollute unless most consumers are knowledgeable and altruistic

Pollution wouldn’t be primarily handed via boycott. Pollution is another thing that would be court-regulated in the libertarian world. For minarchists, this is easy: there is a State doing the regulation. For anarchists, this is harder, but I think it’s still doable. Pollution is one of the hard problems for libertarian theory, and work in that area continues  (Zwolinski, 2014).

Anyway, while boycotts won’t be a main tool for governance in the libertarian world, other institutions can play that role, including general consumer pressure: There are some studies showing that it can induce companies to change their behaviour (Hettige et al. 1996).

1.4 Therefore, externalities justify some regulations

Call them regulations if so you wish. They can also be called definitions of property rights, which are fully consistent -and required by- libertarianism. By excluding fine property rights definitions from the definition of libertarianism, it is possible to create problems for it where there are none.

2. Coordination Problems

2.1 What are coordination problems

Moloch. Or, situations where everyone sees a solution, but there is no coordination to implement it. I criticised the broad idea of pervasive and unsolvable coordination problems elsewhere. Regarding the story of the fish farmers, I copy-paste from myself:

So we have our 1000 firms, each one causing a -1$ loss to others, from an initial gain of $1000 each, so they are making zero. The fishermen get together and lay out their situation. They agree on signing a conditional contract, not too dissimilar from what you can do in Kickstarter: Everyone will pay 300$ for a filter, so that everyone makes $700 a month. The contract will only be binding if everyone signs it, and they go to a court and sign it there so that it becomes binding, and enforceable by an external agency. Now they have changed their incentives: They can choose to do nothing (payoff=0) or sign (expected payoff=700). So they sign. Once they have gone to court, they can either cooperate (expected payoff=700) or can shirk. But now their payoff is not 999, but paying a fine until they cooperate, so now it’s rational to cooperate. Fixed.

But you have used a court!, I’m told. So what? In Libertonia there are courts, what were you expecting? Third-party enforceability is a valuable service.

But what if there’s a stubborn fisherman who demands others to pay him to actually sign the contract. There we run into the question of whether the fishermen actually had the right to pollute the water in the first place.

So suppose that the fishermen raise this issue. Most of them argue that since the lake is a common resource, and since it’s in the interest in all of them for it to be clean, they should regard it as a common property, where rules can be imposed to ensure inner coordination. The hypothetical free rider could argue that since he was the first to get to the river, he has the right to pollute, and that others ought to pay him. This could be some sort of Coasean bargain. In this case, if he demands payment P from each other, he would be making 700+999P. Others would be making 700-P. In the cooperative case, total earnings are 700*1000=0.7 megadollars. In the noncooperative case, 700+999P+700*999-999P=0.7 megadollars, same amount, different distribution. P would depend here on a variety of factors, but total gains are the same. The court system would have to deal with cases like this, and resolve disputes, set precedents, trying to ensure good outcomes while respecting rights. No trivial task, but not an impossible one either.

Read also this. On fisheries, In general, read  Ostrom (1999).

The Coasean solution is not troublefree: it is necessarily efficient only if, at least, there are no transaction costs. Real life bargaining will not be texbooklike efficient. But as is highlighted through this FAQ, these are not the only solutions. David Friedman discusses here Coase’s Theorem.

2.1.1 But this doesn’t work

Empirically, yes. Read Leal (1998). The FAQ proposes fishing quotas. The obvious libertarian solution is private or common property rights, to which fishing quotas are an approximation. On a comparison between private and common property rights, see Ostrom and Hess (2007).

The example of the Atlantic Northwest cod fishery was an example of government failure: the fishery was government (mis)regulated. What would have happened had there been property rights on the whole fishery? The government didn’t allow that to happen. The US fishery policy has its problems (White, 2000).The EU’s Common Fisheries Policy also seems (Khallilian et al, 2010) to be failing. Governments can succeed or fail at solving common resource coordination problems, but so can private or communal institutions (Tarko, 2012).

That said, the state of global fisheries today is dim, even with plenty of national and international government regulation. Creating common property in fisheries that span national borders conflict with sovereignty (Ha, mere fishermen telling States what to do with their territorial waters!), so in a way States could be making their governance more difficult.

2.1.2 But Fishermen won’t do it!

As said above, they actually, empirically, do it, especially for non-international water regions, and when rights of some sort are in place. Formal external enforcement will be a feature of any libertarian political order, with or without a state, so the fact that it is required is no critique of libertarianism.

For international waters, things are trickier, to the point that, as pointed out above, States are not able to deal with it (right now). But would we go as far as claiming it is impossible?

2.2 Global warming is an example of this problem, as is ozone layer depletion and recycling

The same problem that affects the private sector affects States. So this kind of problems are not quite problematic for libertarianism qua libertarianism, as States also can’t fix them on a global scale. The statist solution would be to have a world-State to regulate these matters. The libertarian solution would be to say that since everyone makes use of the atmosphere, and that for the purposes needed (regulate emissions of certain gases and pollutants) it can’t be subdivided, it has to be a single property. So it would be justified for it to exist a worldwide atmospheric regulation committee to solve this tragedy of the commons. Sounds really hard to do, but States haven’t been doing that much better lately.

Fortunately, it seems that technology can do what individual incentives are not able to in a world of Nation-States. Right now, we’re living through a solar energy revolution (and if we’re a bit optimistic, we’ll live a fusion energy revolution in some decades) that will end up with fossil fuel energies replaced by a cheaper and cleaner one. So while the problem is not solved, it stops being an actual problem, so everything is ok.

2.3 Libertarians say boycotts can discipline corporations. Incentives aren’t aligned that way.

Suppose markets are more competitive so that the difference between one product and its replacement is slim. Then it’s maybe $1 plus an ethical cost of buying from an unethical company vs $(1+x) and buying from a nice company. If x is relatively small, it may work.

Now, I’m not arguing boycotts are a really really strong tool. But those unspeakable things can be solved by courts. I mean, if there is a company that uses slave labour, anyone could sue them for slavery, on behalf of the slaves. I don’t really see the problem here.

2.3.1 They don’t work: the Coke example

Those kind of abuses are highly rare, and they also happen with governments around. Governments that, in some cases, are corrupt. Some left-libertarians even claim they happen because there are governments around.

And the Coke example doesn’t work that well. It wasn’t Coca-Cola itself, it was a bottler that worked for Coca-Cola, the evidence for the alleged crime is dubious, and the lawsuit ended up dismissed.

2.3.2 Governments can do better

Most of the laws mentioned had little to do with people wanting things. Except for very particular cases, there is a notable disconnect between what people want and what the government does in its day to day activities. Overall, governments, and any institutional system, anarchies included, are constrained by generally accepted beliefs about what’s right and wrong (Long, 2008, Wiśniweski 2014 ). But within this, governments have lots of leeway.

Some of these regulations could be actually said to privilege the corporations they regulate.

2.4 Coordination failures also justify government spending on charitable causes

Regarding world hunger, it is not that clear that there is a coordination failure here. Why? Depending on whether people value alleviating people’s suffering vs feeling good by making donations, it will or won’t be. For the first case, it is a private good. For the second one, it’s a public good. In the first case, for each dollar you donate to a charitable organisation, you get a marginal reduction of poverty or hunger. It’s not that if everyone except one person donates, then hunger doesn’t go away. People don’t really care that much about world hunger in practice (Leeson, 2014) . This issue is serious, yet how many people devote time to inform themselves about how to better deal with it (e.g. Effective Altruism, knowing about Givewell, etc)? Almost no one. Yet those resources to acquire information are free.

In the OECD, which I will use as proxy for developed countries, there are about 1.3 billion people. Let’s assume all of them can donate. So 1.3*1*(365/4)=118.6 bn$ per year. If Jeffrey Sachs is right, and ending poverty costs 175bn$ per year, people would have to pay 1,5$ per week for 20 years (This assumes kids will also pay, so the estimate is a lower bound). But Sachs is not without his critics. Givewell accepts that sort of criticism for that sort of aid plans, while defending that their approach: bottom-up, cost-effective, step-by-step interventions doesn’t have the same problems. Sadly, there are no obvious solutions to lift the world’s poor out of their situation, and if there isn’t, I think it’s difficult to justify coercion to help towards that goal.  So far, $4.6 trillion have been donated towards foreign aid in the last 50 years. Is it really believable that just with 175bn$ per year for 20 years, there is going to be a substantial increase on the global quality of life of the poor, above that produced by economic growth?

Finally, if this argument succeeded, it may be that it proves too much: Wouldn’t we have to, on consequentialist grounds, eliminate the welfare State to a bare minimum, and donate huge amounts of money to the absolutely poor?

If people really cared that much, they have a handy mechanism to make everyone pay: the State. Massive foreign redistribution, or open borders is something that’s hard to see defended by political parties, or asked for by voters.

One further thought is that States may be making this problem worse. The EU’s Common Agricultural Policy, and US subsidies for biofuels do harm the poor in other countries, while benefitting interest groups inside their borders.

2.5 & 2.5.1 We need labor regulation 

We don’t. Outsourced to Bryan Caplan.

3. Irrational Choices

3.1 People are not rational. They don’t choose the best means to reach their goals: if someone is signed up into a pension plan by default, he will stay. If someone is not signed out by default, he won’t join it. Transaction costs are almost zero, so it shouldn’t make a difference.

Governments are also people, and coercing people to implement alternative choice architectures to fix minor cases of irrationality that exist are not justified, especially if the proposed solution is worse than the problem. Highly problematic forms of irrationality will end up inducing social norms and institutions that will fix them.

The arguments here seems to be like

1) Markets maximize welfare under stringent epistemic conditions including apt probability estimation, stability of preferences, an absence of framing effects or externalities, etc.

2) One or more of those conditions do not hold in a given market because individuals are less than fully rational.

3) That market thus fails to maximize welfare.  (Boettke et al. 2013) .

But markets don’t really need that to be efficient.

Consider the case of pensions. If one behavioural argument is right, then people won’t save enough for retirement, and we’ll have masses of relatively poor elderly people. This hypothesis is grounded in human psychology, and presumably, for the argument to take effect and be immune to the critique of the welfare-induced irrationality (“The ultimate result of shielding men from the effects of their folly, is to fill the world with fools” (Spencer, in Zwolinski 2015)), we would have to observe those masses of people in really bad conditions before the coming of public pensions. Also, public pensions will have appeared because private pensions were not working, or existed but were stagnant, or something like that.

So why did public pensions appear? There are many theories, and a paper that studies them, and finds that several reasons account for it. The first countries in adopting them were Germany,(1889), Denmark (1891) and New Zeland (1898), and the last ones were Switzerland, (1948), Netherlands,  (1947), Argentina, (1944), and Japan (1942). (Cutler et al. 2004).He doesn’t mention the terrible conditions of the old as a general, or even prevalent explanation, so the hypothesis of massive market failure in providing old-age pensions is in principle rejected. One question would remain: what was the poverty rate for old people before and after mandatory pension schemes?

Before public pensions there were sickness funds. Non-profit associations for the provision of , amongst others, old-age pensions. This was a time before financial markets were well developed, so there was no possibility of widespread pension funds as we know them today, and yet people managed to avoid ruin, in general. (Immergut, 1992). Ideally, we would look for statistics from before and after mandatory insurance was introduced, but those a hard to come by.

After some Google Scholar archaeology, I found some statistics from the pre-Social Security period in the US, and I think this confirms my account more than Scott’s: the elderly do fine without State intervention.

First, there’s Weaver (1987) who explains how society cared about its elders before the State took care of them. There were, as described above, several mechanisms, including savings, family care, various types of insurance, pensions, and charity, and the reason why Social Security appeared was not due to a systemic problem, but due to the harsh conditions imposed on the poor by the Great Depression. Instead of being a set of emergency measures, Social Security was instituted as permanent, and then grew to be what it is now (Miron and Neil, 1998). Some even argue not only that these institutions were good enough, but that they were economically optimal (Emery, 2010)

They worked until a later age than now, too. But we also know that retirement age had been going down decades before Social Security, and continued going down at the same pace after it (Lee, 2000a)

Papers that do endorse public interventions, like Valdés-Prieto (2004) are more theoretical than empirical, and don’t take into account the historical evidence like Carter and Sutch (1995), or Gratton (1997), who shows that the Social Security Administration overreported elderly poverty in order to have legislation passed to strengthen their role.

Earlier scholarship on Social Security did show that the elderly were impoverished, but now the consensus view is that they were not, though some disagreements remain (Lee, 2000b).


Until convincing evidence proves otherwise, we should assume that most older people in the early twentieth century fared relatively well by the standards of the time, that the economic well-being of the elderly improved during this period, and that younger people expected to achieve economic security in old age. Given these findings, the creation of a welfare state dedicated to generous transfers to the aged becomes a great curiosity. Nothing in the economic evidence substantiates overwhelming need among the elderly, as opposed to some other group in U.S. society. Indeed, by the late 1930s, SSB bureaucrats knew full well that children were much more likely to be impoverished than the aged. What then was the source of Social Security? Weaver’s political economy approach identifies a state bureaucracy intent on expansion of the old age insurance title in the social security program. (Gratton,  1996)

3.1.1 People’s preferences are complex, and are not (fully) revealed when they act.


3.2 Many choices are not really choices, but spontaneous, unthoughtful, reactions that don’t conform to rational principles. There is an unequal distribution of abilities to take those irrationalities into account: not everyone is a perfect reasoner übermensch, so some paternalism is justified.

This is true. But you need an extra argument to go from “we are not equal in X dimension” to “It is morally right to use coercion as to reduce inequalities in X”

3.2.1 People will be happier and have more preferences satisfied if they don’t make irrational choices. More rational choices will improve also the economy.

I haven’t seen evidence of the economic-growth enhancing effect of behavioural economic policy. But while it could be granted that coercing people to design their choice architectures so that the default choice is the rational one may improve preference satisfaction, it violates the preference of the coercee not to be coerced. You could just have a nonprofit and tell people to use nicer default options, if the costs of changing default options is minimal.

3.2.2 Predictably irrational behavior justifies government pensions, consumer safety and labour regulations, advertising regulations, concerns about addictive drugs, and public health promotion.

Not really. You may say, well, but aren’t people not saving for their retirement a hugely bad thing? Surely you don’t want to have most of our elders starving and dying en masse. No, I don’t. We can solve this issue by looking at how this used to work before the welfare state, and old people didn’t die en masse. Why? Perhaps because you knew that if you didn’t save, you were going to die, or be a burden to your family and friends, and you don’t want that. That ended up crystallising into social norms ad institutions.  Nowadays, you need not worry about that, and so there’s less incentive to worry.

It may justify some things, related to harming others mostly. All of them implementable within the libertarian framework. And it may also justify limiting the State (behavioural public choice) and even welfare. See also Murphy (2015) for how irrationality affects different political arrangements, even considering politicians are more rational than the average voter.

4. Lack of information

4.1 What is means by lack of information?

There is no perfect information, acquiring it is costly. People usually want some information about safety, efficacy, and the ethics of how a product is produced. Governments can listen to these demands and enforce them, and markets can do so by directly catering to consumer’s demands.

4.1.1 People won’t spend time to know about safe and effective products

Won’t they? In some instances they won’t, like with sugary products. But in general there are mechanisms to easily get information: First, the envelope of the product itself tells things about it, as does its price. There may be third party certifications, consumer websites with information, or even just word of mouth. Ultimately, though, selling really ineffective and unsafe products is a case of fraud, and we know what happens to fraudsters in Libertonia. In the particular case of sugary drinks, I think right now a norm is spreading to be more cautious about their consumption.

What about credit rating agencies then? Weren’t they selling defective merchandise (fake ratings)?  Yes, but it wasn’t because of any fault in the market mechanism, but because of government mis-(not under)-regulation. (White, 2010White, 2010b, Cole and Cooley, 2014)

4.2 [[Insert list of examples of unethical corporate behaviour]]

4.2.1 Not even smart, educated people know about the things described in the examples above

I don’t have much experience with those products, which may explain why I have little experience with them. Just think about this: if something was so bad that clearly didn’t work, people would stop buying it. If something is so unsafe or it is made violating someone’s rights, it should be possible to sue the company over that.

At this point, I’ve replied to a few question by pointing out to courts, and you may feel that this is cheating. It isn’t: Delineating property rights require, among other things, to determine a threshold for when a risky action is considered a rights violation even before the actual damage has happened, and this cannot be done a priori. This is the fundamental moral principle upon which libertarian market regulation would rest.

4.2.2 I won’t tell you which of the examples were true or false


4.2.3 If you believe you care less about product safety because of the government, you’re ten times more statist than me!

Well, not necessarily. You can believe governments do a good enough job in this area, and not bother with really doing your research.

4.3 Without consumer regulation, small business could be destroyed, as only big companies would have an incentive to maintain their reputation by selling good products.

For one, this isn’t necessarily a bad thing. It just means you’d have a different market structure doing the job, but the products would get sold and bought in the same way.

This ignores the possibility of a) Small business owners not being jerks b) Suing them c) Third-party certification

4.4 This justifies consumer and safety regulations

I think it doesn’t. Plus you can do them without taxes. Here is how:

Imagine you have a new product you want to bring into the market. In the current system, there are specifications that tell how you have to make your product so that it is safe. Additionally, you may have to first prove to a governmental agency that your product is safe. Then, you can sell it.

In an alternative system, there are known standards (courts!) that the products have to comply with. Say, computers must not electrically shock users, food must be in good condition, cars shall not spontaneously explode and so forth. Companies then are free to do as they wish to make sure their products pass the requirements. Then, they stamp a logo on their products saying that it complies, and sell them, without having the government say anything, or maybe just notifying them that they are selling the product. Then, from time to time, governments or private agencies go to the market, buy some products, and see if they comply. If they do, ok. If they don’t, the company is sued for massive fraud.

What do you think of this second system? It seems workable within a purely private setting. And what if I tell you that something like this is the system that is in place in the European Union since 1985, replacing the system described in the first paragraph? Okay, it’s not a private system. The EU makes the rules and governments enforce them. But it’s closer to what a market solution would look like.

Regarding regulatory agencies more broadly, there are reasons to be sceptical of whether they do more good than harm, see this for the FDA, and this quote from the Oxford Handbook of the Economics of the Biopharmaceutical Industry

“Despite the central role of the FDA in regulating quality and R&D costs of medical products, economists have conducted relatively little theoretical or empirical research on the efficiency of FDA policies. Ironically, if a product application were presented to the FDA with the scant amount of evidence that currently exists on the efficiency of the policies of the agency itself, such an application would likely be rejected on the basis of insufficient evidence. (Malani et al. 2012)

Part B: Social Issues

The Argument: Most of the difference in success are not attributable to hard work or intelligence, but to trivial, unchosen, factors. Therefore, most of people’s lives is determined by unfair factors.

If you think trickle-down economics works to benefit the poor, you are deluded, it doesn’t work, as history proves.

The Counterargument: That’s basically true. Except that fairness does not univocally determine justice or entitlement. That’s why the argument fails.

Trickle-down economics has never been defended by any serious economist, and it was born as a strawman.

5. Just Desserts and Social Mobility

5.1 Children of the rich are more likely to be rich, and children of the poor are more likely to be poor

True, but maybe not to the extent the FAQ says. If intergenerational income elasticity means, roughly, how much mobility there is, Denmark is three times more mobile. If instead of measuring intergenerational income mobility, you measure rank-mobility (Correlation between parents being in a quartile and the quartile the child ends up), the US has an average upward social mobility (and greater downward mobility), on par with Europe’s (Corak et al. 2014). If you instead focus on how many people born poor stay poor, the results do show that the US is less mobile: around 66.6% (or  58.2%, in another measure) of the poor remain poor,  in Denmark this figure is 60.5% (Compare 62.5% for UK,, 71.9% for Finland, 62.5% in Australia). If you then use transition matrices, then you get different results, which confirm again that the US is less mobile, but not as little mobile as elasticities suggest: the US is 10% less mobile than Denmark, not 300%. This isn’t the only issue, though: being poor in the US is worse than being poor in Denmark.

The US is no libertarian paradise, so this isn’t hardly useful for the debate, though. Rather, compare the Nordics to Switzerland, Singapore, Hong Kong or New Zeland. From a Rawlsian point of view, you may prefer to be born in one of these countries rather than in a Social Democracy. As an example, if you measure Eurostat’s material deprivation rate by age group in 2013, you get that in Denmark 8.8% were poor, compared to 3.7% in Switzerland, 11.6% in Germany, 8.5% in Finland, 4.3% in Sweden, 5.6% in Luxembourg, 4.8% in Norway, and 17.4% in the UK, and 16.9% in Spain. Granted, in all of those countries there is a State that does redistribution, but their sizes wildly differ. Given a desired outcome, it seems reasonable that the less coercion required to achieve it, the better.

That said, I don’t know any academic libertarian philosopher who talks about this moocher/productive distinction, or that hard work makes you deserve stuff.

The US is also an argument why having the State spend more does not necessarily translate into better outcomes for the poor: the US has a larger welfare state than Denmark if you account for tax breaks and other things. Where’s the catch? Denmark focuses more on the poor, and the US more in the middle class.

To sum up: The Nordics are very good, but not optimal, nor an argument against libertarianism. Some libertarian authors actually argue that the Nordics are more libertarian than the US itself!

5.1.1 But the conventional knowledge among progressives is different!

It’s because it is based in a different methodology: measuring intergenerational elasticities of income. We can ask different questions: Do children of poor parents have better prospects in Denmark than the US? Is there a higher correlation between children and parent income in the US than in Denmark? Is it easier to escape poverty in the US or in Denmark? Welfare-wise, the question to ask is the first one. And prima facie, in Denmark they do better, but not because of a lower intergenerational income elasticity, but because of freer markets, and a better designed welfare State.

You could now ask: Well, if you think the Nordics are so libertarian vis-a-vis the US, if you had a magic button to transform the US into a huge version of Denmark (Copying the best things Denmark has to offer, so that the end result is a mix of the best of both worlds), would you press it? And my answer is yes, I would. I would also press a similar button to transform the US into Switzerland over the Denmark button, and the minarchist button over the Switzerland button. And if we’re ex-ante sure it works as its theorist say, then the anarchy button is at the top of the ranking.

5.1.2 Things like genetics, and parents teaching better habits could explain part of the wealth gap, instead of parents just handing money to children


5.1.3 Social mobility is a political choice, not something given. People are naturally unequal.

True, but the data shown before shows that welfare-wise, the US doesn’t do it that bad compared to Europe. Also, some of the countries that have lower income elasticities than the US have freer markets than the US (as measured by Heritage’s Index of Economic Freedom), especially if you discount the welfare State. You can have both a free market and a welfare State, and a less free market and less welfare, so that the former is globally freer than the latter, depending on how the index is constructed.

Plus, the Nordics were already egalitarian, high social mobility Nordics before it was cool they had a welfare state. And Nordics living in the US enjoy much better prospects than the rest of Americans, which calls into question whether the Nordic success is just an institutional matter that you can easily copypaste elsewhere. The government can improve social mobility

Scott cites Mayer & Loppo (2008) Well, maybe. Data shown above casts some doubt on this, but prima facie it is highly plausible that a welfare state will improve the prospects of the poor. (How much, 10%? 500%?)

But, in the long term, it may have perverse effect, such as incentivising the same behaviours that keep people in poverty. This is not blaming the poor for their poverty, just recognising a possible cause and effect relation. And then, if Gregory Clark is right, social mobility (broadly understood) is more or less fixed, so you can only aim at improve the conditions of the poor.

Finally, a declared aim to improve the lot of the poor may not actually do so, but end up extracting resources to give them to the middle classes in a greater proportion than to the poor: what is known as Director’s Law (Mohl et al., 2008).

5.2, 5.2.1 Hard-working and conscientiousness, and so on depend on external factors

True. If you take into account parents, genetics, and so on, what one is able to do above that is generally not that much.

Regarding the free will debate, if there is no free will, it’s hard to see how you can have merit or desert, except perhaps via compatibilism. But if there is, merit and desert aren’t relevant philosophical concepts either, as I’ve argued elsewhere, so our argumentations shouldn’t hinge on that.

5.2.2 Lead poisoning reduces IQ, self-control and other sort of bad consequences. Is it fair to blame someone for being less smart due to higher lead exposure?

True, and it is not fair to blame them. And phasing out, and finally banning, leaded gasoline, was correct.

Lead is an interesting example. It was assessed and allowed by the government back then, and when evidence arose that it was dangerous, industry accepted, voluntarily, to restrict the use of lead as an additive, and complied with it, before the EPA formed, and ended up banning it.

5.3 Therefore, if success is externally determined, redistribution.

Wait, why?

I think the FAQ has the same problem Rawls had: He notes that no one deserves anything (more or less), and from there he follows that everything is up for grabs. As I argued in another post, this argument also licenses, in principle, organ redistribution, and thought-engineering to ensure people act in such a way as to be people that work towards a society that makes the worse off better off, want it or not.

Here is the Rawlsian view:

Since most of who we are and what we do is greatly influenced by undeserved native endowments and by the undeserved circumstances into which we are born, one cannot deserve anything, or, at best, one can deserve very little.  According to a common interpretation, Rawls believes that desert should not have any role in distributive justice, since these undeserved factors have a major influence on all would-be desert bases (Sher 1987, 22 ff).  (IEP)

The problem is not rejecting desert. I basically reject it. The problem is saying that desert is a precondition of entitlement, and that requires an argument.

Rejection of desert or merit, I think, is the mainstream philosophical view today, including among libertarian academics. And even if it were not, the mainstream view among libertarians is that there is no connection between desert and entitlement. It was Robert Nozick who famously said:

We feel more comfortable upholding the justice of an entitlement system if most of the transfers under it are done for reasons. This does not mean necessarily that all deserve what holdings they receive.


We noted earlier that the entitlement conception of justice in holdings, not being a patterned conception of justice, does not accept distribution in accordance with moral desert either. Any person may give to anyone else any holding he is entitled to, independently of whether the recipient morally deserves to be the recipient. (Anarchy, State, and Utopia)

Hayek and Milton Friedman also rejected that.

To be sure, there are libertarians who accept the notion of desert and merit, like Bryan Caplan, but they don’t make desert the core of their justification for free markets.

This mistake is what I’ve called the Unjust World Fallacy. The well-known Just World Fallacy, applied to this case says: 1. Poor people deserve being poor 2. If they deserve so, it is just for them to be poor 3. Therefore, it is just for the poor to be poor.

The Unjust World Fallacy is: 1. Poor people don’t deserve being poor 2. If they don deserve so, it is unjust for them to be poor 3. Therefore it is unjust for the poor to be poor.

Both second premises are mistaken, as there is no justice and desert connection. Additional arguments are required to go from lack of desert to lack of entitlement and the need of redistribution.

That said, you can defend that there exists a general moral duty to help the poor, but not because they deserve it, but just because it is good to help others, and doing more good is better than doing less good, so if you are going to dedicate resources to helping others, if you target the poor you will get lots of helping done.

6. Taxation

6.1 See the Moral Issues section for a discussion of whether taxation is inherently evil

See my reply to the above in the corresponding section

6.2 Progressive taxation is not unfair: it is utility-wise equal taxation. Flat taxes are unequal taxes with respect to the value people give to the money they have

You can still have flat taxes and equal burden: As pointed out by David Friedman, if there is a person A with 40,000$ and a marginal utility of income of 2 utiles/dollar, and a person B with 80,000$ and a marginal utility of income of 1 util/dollar, a flat tax of 10% will get 4,000$(8,000 util) from A and 8,000$ (8,000 util) from B.

Also, if this argument goes through, it doesn’t justify your typical redistribution. It justifies taxing everyone progressively in developed countries and throwing the money at a vastly enlarged version of Givewell.

Read White (2015) for debunkings of economically based arguments for redistribution. Redistribution, to be justified, needs moral arguments.

6.2.1 Progressive tax systems don’t work the way some people think


6.3 By what standard are taxes too high?

By historical standards

6.3.1 Historical standards show the size of government is more or less stable.

Historical standards show the government has been steadily growing, and people have to pay more and more taxes. We should look at the fraction of social income that is extracted by government

Here there is a chart of taxes and expenditures from the year 1900 in the US. Expenditures more or less track taxes, so it’s a good enough proxy.


Here you have longer term data from the Concise Encyclopedia of Economics


Government expenditures over GDP (USA)


Government expenditures over GDP (Sweden) Taxes for the rich are the lowest they’ve been for the past 75 years

Technically, yes. But in the longer term, no. (I know these are not effective rates, but I couldn’t find those. You get the idea, anyway)

highest-marginal-tax-rates-1913-2013 The rich are getting richer and the poor, poorer, and as more money concentrates in their hands, more of the taxes come from them

The poor are not getting poorer, after tax incomes are growing for everyone, at diverse paces. Tax rates for corporations are also really low

Not only they are high historically, in the US they are absurdly high even if you compare them to, say, the Nordics. Estimates for the effective tax rate vary, some say it has been going down, so that they are somewhat lower than OECD average (16.1). Corporations in the SP500, though, pay around 30% of their income, which is really high. So the companies you think of when you think of ‘corporations’ do pay really high taxes.

6.3.2 We’re at the left of the Laffer curve, and it is false that high tax rates discourage production, real income has low elasticity with respect to taxes. What taxes affect is taxable income, so increasing taxes x% will bring less than x% of revenue, but it’s still positive.

If overall taxation is a proxy for government size, then unfortunately the consensus is that increasing the government size (and taxation required to fund it) by 10 percentage pints leads to a 0.5%-1% less lower growth rate (Bergh et al. 2011, Bergh et al. 2015).

If you measure aggregate wages and aggregate productivity, there has been no decoupling, if you were thinking in that.

6.4 Trickle down doesn’t work

As said before, no serious person defends trickle-down.

Are the poor getting poorer? You need to look at full compensation, not as just income. Market income depends upon what the government does, you can’t just measure it in isolation. As wages=productivity more or less holds, you can estimate that free market wages=actual compensation today. Doing that, we see that the poor are not getting poorer, but richer, at a slower pace than the rich.

6.5 Raising taxes won’t fix the deficit. But cutting lots of things, including all of welfare, won’t do it either.

US federal deficit was 438$bn in 2015 and 616$bn in 2016. Let’s say 500$bn. When Scott wrote, it was 1.2$tn. If Medicaid, Medicare, and Social Security are cut, and defense is cut in half, you get savings of  1.9tn$, so technically, you can do it. Now, you wouldn’t want to do it in a year, that’d be dangerous. For the case of Spain, there is a transition plan proposed here to a minimal state (5% of GDP), and it takes 50 years to transition.

6.6 Little taxes go to foreigners and the poor. Most taxes go to benefit middle income citizens.

Indeed. Welfare states are not about rich-to-poor redistribution, but mostly medium class-to-medium class redistribution. Nordics included.

This is because of Director’s Law (also here, in Spanish).

Part C: Political Issues

The Argument: Governments sometimes do things right, and there are reasons people are biased to think otherwise. Government-run healthcare systems are better than private systems, and some libertarian proposals like prison privatisation are insane.

The Counterargument: Yes, governments sometimes do things right. However, private healthcare does work better than government-run healthcare systems. Private prisons in the US aren’t a good example of private institutions.

7. Competence of Government

7.1 Government landed a man on the moon

[[Insert pic of alternative uses of 170 billion 2005 dollars, and the engineers and scientists dedicated to it]]

I say this as someone who particularly loves space and rocketry, and who sees the landing on the moon as one of the crowning achievements of mankind.

7.1.1 If that’s not enough, governments also brought us: smallpox and polio eradication, worldwide. Cholera and malaria eradication in the US, the computer, the mouse, digital camera, and email. The internet, and regular highways. Clean, free water, and electricity to the whole continent. Forcing integration and leading the struggle for civil rights. GPS. Accurate disaster forecasts for a series of disasters. No bank runs. Nuclear power and game theory.

I think we’re not in agreement here.

  • Smallpox and polio eradication. Okay for Smallpox. Polio eradication is a public-private effort. Other illnesses have been targeted and successfully reduced by nongovernmental-led efforts (Dracunculiasis, Hookworm ). Ongoing efforts against malaria are public-private.
  • Cholera and malaria eradication in the US. Okay.
  • The computer: Advances in computing have been happening since Babbage, and important advances like Zuse’s were initially self-financed. That said, yes initially with cryptography and military requirements for computation, government played an immense role.
  • The mouse: Doug Engelbart (and some guys at Telefunken) is to credit for that one. He received government funds for it.
  • Digital camera: Invented at Dresden Eastman Kodak
  • Email: If you mean AUTODIN, it was designed by RCA, IBM and Western Union (for the Air Force). If you mean host-based or LAN mail systems, there were a few private ones. Or maybe you mean email networks, in which case it was ARPANet’s mailing system, see below.
  • The internet: Modems, glass fibers, most communication satellites, almost all submarine cables (which carry 99% of the traffic), ISPs and their peerings, and every website of interest, is privately designed, owned, and managed. That is the internet. Now, if you go to the network that preceded the internet, ARPANET, it was the work of ARPA, and a few subcontractors. I have no trouble granting the invention of the internet to ARPA. Read here a story.
  • Regular highways: Private highways precede government’s, as do private roads, in general. If you mean building more of them, then yes, the government did that.
  • Clean, free water: Work initially began by private utilities, plus it isn’t really free.
  • Electricity: Same as above.
  • Forcing integration, leading the struggle for civil rights: Even if this were true, you would have to weight this against flagrant violation of civil rights (think of apartheid in South Africa or segregation in the US). Regarding civil rights, I don’t know what Scott exactly means, but I think civil rights extension was a thing of civil rights movements, not governments.
  • GPS: Okay
  • Disaster forecasts: Okay. Governments have been at the satellite business from the beginning, and competing with free services is hard.
  • No bank runs: Okay. But bank runs before were usually caused by government regulations, and the possibility of them happening is a good thing and government management of money has been worse than what came before it (Selgin et al. 2012), and even worse when compared to fully private banking. Inflation under the FED has been higher and more volatile (Hogan, 2015). The banking system ought to be reformed.
  • Nuclear power: Okay. But you can imagine an alternate future where nuclear energy progressed at a slower pace, with safer, and smaller reactors, so that it gets less bad press. The general problem with innovations is that it’s hard to know what would have happened if some particular agent hadn’t player it’s role. My position on this is that for technological improvement it doesn’t matter, and that for practical implementation, it does.
  • Game theory to avoid misusing it. Game theory was birthed by von Neumann, without government intervention. After he invented it, the government subsequently hired him to continue his work.

Whatever I could had said above, what matters is whether the private sector would have come up with those inventions and advances, and I see no reason why not. We could then debate, case by case, how longer would have taken for a market process to achieve them.

Also, think about the many things governments did wrong or that could have done wrong (nuclear war). Right now, think about over-regulation and excessive defense spending in the US. Governments brought peace, plus some links

Or maybe guns did. Or moral progress. I don’t think the Hobbesian logic works (see Huemer’s book for an explanation). Regarding the links, I’m going to skip doing a debunking, I don’t think it would add that much substance to this FAQ at this point.

7.2 Large government projects are not more bloated and over-budget than private ones, according to one study

Scott links a paper that studies costs in transport infrastructure. The critique here is not that project building will be more expensive, but that it will be less economically rational vs. the market.

Also, I want to remark that the study applies to transport infrastructure, not large projects in general.

7.3 Some of the world’s greatest companies are or have been state-run like JNR or the BBC. Sometimes, those corporations are set up to provide goods the market wouldn’t provide

Okay. I won’t go deep on Amtrak or the USPO, but I think it’s difficult to defend those on that basis. If it is more efficient for small urban concentrations to have more delays in their mail deliveries, so be it. Incentive for them to move to larger cities. Amtrak should also go the same way. If people actually want passenger transport, they will pay for it. One cause of Amtrak’s problems might have been an oversupply of public highways.

7.3.1 Advances made by the government included what was mentioned above, plus the radar, jet engine, satellites, fiber optics, artificial limbs, and nuclear energy, plus at government institutions. Many market innovations come from de-facto state-owned monopolies, like Bell Labs. Apple does consumer packaging. De novo invention comes from very large organizations that can afford basic research.

I commented already on some of those advances.

  • Jet engines: I wrote about that here, and here. Not invented by government, but improved after invention by it.
  • Radar: Not invented by government, but independently discovered and improved by it thereafter
  • Satellites: Okay
  • Fiber optics: The key achievement was made by Corning Glass Works. Government’s role was relatively small here.
  • Artificial limbs: In recent times (post WWII), there has been a lot of government work here. But there was work in that field before that.
  • Nuclear energy: okay

Contra Scott -and Schumpeter- innovation does not happen just at large companies shielded from competition. There is no causality pointing from more concentrated markets to more innovative markets. Other researchers find an U relationship, where the largest and smallest firms are the most innovative (Cohen, 2010 [sec 2.3]). If we use R&D Magazine’s 100 Awards as proxy, we find that most innovations come from the private sector, and that, unlike in the past, they come usually from small companies. Where breakthrough discoveries (not inventions) are made is in basic-science research centers, and those are mostly universities, and yes, big corporations. And universities are heavily government funded. But they need not be.

Regarding Apple, what Scott says is true. He doesn’t claim the US government is responsible for the iPhone, but there are people who do. Here is a reply to that.

7.4 The fact that it seems that many government programs are failures is based on a biased perception.

Okay. Negativity bias is strong. The same thing explains why there are people who think the world is getting worse.

7.4.1 Many things governments do are not known, so its contributions are understated


7.4.2 Yes, sometimes regulations don’t cause better outcomes, things were improving before them. But apply your skepticism fairly and evenly. It isn’t the case that government destroys what it touches, but it might be that even if what it touches goes well, it’s not because of the government.


7.4.3 Regulations foster social norms, like wearing seatbelts, perhaps without things like that there would be less lifesaving regulations

Perhaps. But take into account that mandatory seatbelts encourage less careful driving, and hence more deaths. Lives saved by mandatory seatbelts may equal lives lost due to them. Nonprofits can also play that norm creating role.

Also, the rates of usage were rising before they were made mandatory.

8. Healthcare

8.1 Government-run health systems do better than private health systems, and are cheaper. Sweden, France, Canada UK (single payer)  vs the US (private). Public systems win at life expectancy, infant mortality, and cancer.

I reject the fundamental premise: the US does not have a private healthcare system. Or, if we accept that it does, it is private in the sense the Federal Reserve is a private bank. One thing is that a substantial part of the US healthcare system is privately owned. Another is the degree of regulation that sector has, and the effects it has over how it behaves. I will explain this in this section.

Another criticism to be made is that the populations are not the same, and that biases outcome comparisons in the variables presented by the FAQ. See here for more.

No one defends the US healthcare system as it is now, as far as I know. And it’s not because of an implicit no-true-Scotsman fallacy: there are reasons why libertarians criticise the US healthcare reason, and propose measures to fix it:

The problem of US healthcare in a nutshell: In a healthcare system, someone should be doing the rationing. In the US, no one is. Unlike in Europe, the public part of the US healthcare system does contribute to price inflation because of that.

Lots of things could be said about healthcare and how a pure free market there would work, but this is a Non-Non-Libertarian FAQ, not a place where to present a full defense of any particular position.

8.2 Government-run healthcare is more efficient than market healthcare: the Blue Cross example, and a chart of spending vs outcomes

The claim of Blue Cross is a  bit hard to evaluate. The chart the FAQ links has no source but some googling leads us to a book by Peter Drucker. Drucker does not cite the source for those numbers. But Blue Cross in Massachusetts covers around 2.8 million people, and the Canadian Health system 27 million. Health Canada has 10403 employees, and BC has 3600. Hence, the claim made is false. It may be then argued that the people/employee ratio is better in Canada, to which I will reply that there are economies of scale, and other factors at play.

The chart presented here is not fair for the reasons presented in 8.1. Instead, you could compare the share of GDP expenditures on healthcare vs GDP, and find out that the US is not that much of an outlier (see also this) depending on how you measure it, plus that costs have been increasing also in single-payer systems (not as much as in the US, in some cases). But as I will say later, this is to be expected.

8.3 Waiting list are not a big issue. Health has to be rationed. In a private system, the price system does this, but it has the disadvantage that if people want too much healthcare than they really need, prices will spiral up. In contrast, in public systems, allocation is based according to need.

If people want to see a doctor more than they ‘need’ to do so, that is just an increase in demand, and the mechanism described by the FAQ will work equally well. If the system excludes masses of poor people, there are strong incentives to find out alternatives to serve that market.

8.4 Death/life panels 

Fine with that

8.5 Government healthcare is more effective because of economies of scale, and people are irrational and don’t know what healthcare they need. And for-profit hospitals have worse outcomes than non-for profit hospitals.

Economies of scale have little to do with it, beyond the fact that governments can impose their dictates on companies if they are a single buyer (a monopsony). Above a certain threshold, an insurance pool adds little fixed-cost-sharing by adding more people to it, and insurance pools can always opt to reinsure themselves.

Regarding profits/nonprofit hospitals: Fortunately 75% of private US hospitals are non-profits (note that the study doesn’t compare private v public, but this does). The results are not robust (they depend on regions, and time periods), and apparently public hospitals aren’t as good as private non-profits (Eggleston et al., 2008 Koning et al. 2007) And the problem here is not the profit motive: non-profit hospitals ‘owners’ do make ‘profits’ via paying higher salaries to their boardmembers (and other workers)

If you want to look at something that looks closer to the system libertarians imagine, look at Singapore. Yes, there is plenty of government intervention, but libertarians will argue that the way the system works can be replicated entirely with the market mechanism. Read also this by Caplan on free market healthcare.

A healthy healthcare system will have lots of out of pocket payments. Insurance should be that, insurance, to be used only for very rare, high-cost things. We can even do a quick analysis of the impact of public and out of pocket payments on healthcare costs, as measured by the cost of a basket of hospital services (Koechlin et al. 2010). If we do the analysis for a series of countries (Australia, Canada, Finland, France, Italy, Israel, Korea, Portugal, Slovenia, Sweden, and the US) We get these charts (orange line is a fit excluding the US):

This is, of course,no conclusive evidence on its own.

We can also take a look at India’s healthcare system. Why, you may ask.

In many poor countries, healthcare is mostly private (Forsberg et al., 2011) . India in particular is of interest because of the some innovations that are happening right now in their private healthcare system,  that people prefer over the lower quality, but free, public one (Das et al. 2015).

These revolution has come to the point where some Indian hospitals  deliver western-level quality care at a fraction of the cost via both process innovations and technological innovations (PSP4H, 2014), while at the same time helping the poor by using fees from the richer clients to subsidise the poorer ones.

Granted, India is not a wealthy country, and lots of poor people become indebted due to health expenses. But this situation is what has spurred innovation in the healthcare sector, and it is not a situation of poverty that will last forever. Health outcomes are improving at a good pace (Joumard et al., 2015). The system, as you may see from my references, is not perfect. But doesn’t it work better than you expected? Now imagine this in a wealthier country!

To finish this section, I can also point you to this article and podcast by John Cochrane on healthcare reform for the US, and how markets can help making its healthcare system better.

9. Prison Privatization

9.1 Private prisons are worse than regular prisons. They also have an incentive to lobby for stricter sentencing to get more prisoners, and thus more income.

I don’t know about this. But private prisons in the US are more like “contracted-out” prisons than privatised ones. There seem to be people that have evidence for them too. In ancient Athens they didn’t work that badly (D’Amico 2009). And you can also have charity-run prisons (Skarbek 2014).

9.2 It’s ok if libertarians don’t believe in private prisons

Minarchists are fine with both types, if private ones turn out to be really bad. Anarchist libertarians would need private prisons, if imprisonment is to be done. Most theorists, I think, tend to be against the idea of imprisoning people, but of making them work to generate money to compensate those they have aggressed against. Or, perhaps, security agencies would pay for the prisoners to be there.

10. Gun control

10.1 Maybe guns should be allowed, but some restrictions should be in place 

Indeed. How to keep nuclear weapons, and similar stuff out of the streets? For minarchists this is trivial: what weapons are reasonable to have can be decided by the State. For anarchists, it is indeed much harder. The solution would have to be for the court system to  ban unreasonable weapons, and since most people would favour such a measure, it would probably be implemented.

11. Education

11.1 But public schools do better than private ones, and we don’t have statistics from before the rise of public education.

Fortunately, we do have such statistics. For UK (West, 1978) and France(Blum et al., 1985). Reminder: Public education begins in both countries around 1870.

Captura de pantalla de 2016-03-03 15-10-56

Illiteracy rate

Plus, government sponsored public education is a horrible failure in underdeveloped countries (Tooley, 2005) compared to private schools.

The case of Finland: Finland spends per student 5% more compared to Spain, and the difference in the PISA ratings is of 40 points. However, if you compare SES-adjusted ratings between Finnish public and Spanish private schools, you get a difference of just 10 points between Finland and Spain (in favour of Finland). A fairer comparison would involve students in Finland itself, but there are no private schools for such comparison. If you compare Macao and Hong Kong (>90% in private schools) with Finland, these countries win over the Nordic country in PISA (500 vs 485-478 of Finland).

Contrary to Scott, School ownership does not make a difference (also this and this and this) in quality if both schools have the same level of freedom to set their schooling policies, after adjusting for SES.

11.2 Libertarians believe in equality of opportunity, but that requires regulating schools

If by that you mean mandated equality, I think they don’t believe in that. As discussed above, genes, and external factors determine to a great extent your chances in life. By default there is no equality of opportunity or of results.

Regarding the extreme cases Scott discusses, in Libertarialand, some ‘regulation’ would be justified as a means to avoid fraud. Offering things that sound like educational services but aren’t is like selling rotten apples. The thing is, this may not be enough for some people. If you want full equality of opportunity, you need to equalize not only schools, but food intake, parents, universities, friends, and mostly everything you find in your life. This is doable to some extent, especially with the future advent of cheap genetic selection, but the philosophical question remains of whether it is justified to coerce others to accept these more extreme measures, or to coerce others to get resources to implement said measures.

11.3 The government should protect children from their parents

Surely. But this is not something ‘special’. Children have rights, and it is right to protect them: any properly structured organisation that can do that will do. It is a violation of the rights of the child to mistreat it, or even not to give it adequate resources, given the resources of the parents. There are interesting debates about whether libertarianism actually requires licensing parents to have children as a matter of justice.

Part D: Moral Issues

The Argument: Moral systems based only on avoiding force and respecting rights are incomplete, inelegant, counterintuitive, and usually riddled with logical fallacies. A more sophisticated moral system, consequentialism, generates the principles of natural rights and non-initiation of violence as heuristics that can be used to solve coordination problems, but also details under what situations such heuristics no longer apply. Many cases of government intervention are such situations, and so may be moral.

The Counterargument: Consequentialism suffers from the same problems that absolute rights theories suffer from, and lead sometimes to even more unintuitive situations sometimes. As of today, there is no fully coherent ethical system to make decisions in any given circumstance, but we can rely on reflective equilibrium to make our ethical intuitions and ethical principles cohere.

12. Moral Systems

12.1 Freedom is not infinitely valuable

Indeed it isn’t.

12.1.1 Franklin didn’t mean what you think he meant


12.2 Taxation is theft, yes, but everyone accepts that. The only force this argument has is emotional

I don’t know if everyone accepts that, but everyone accepts that theft ought not be done in general, and that the particular kinds of theft governments undertake would also be criticised if regular people did them. People think governments have political authority, that’s why they accept it. The argument against taxes has force insofar as regular theft is rejected, which it is.

Strategically, saying that taxation is theft, or that the non aggression principle is true is not useful and shouldn’t be done, because people won’t take it as a starting point.

12.3 Why should the initiation of force against innocent people be wrong?

Most people would say it is wrong, and seems obvious. Consider this thought experiment: You are told that a government is about to use force against an innocent person. You can make that use of force happen or not. Most people would stop that use of force. A consequentialist, given no more information, would be indifferent and flip a coin to decide. Alternatively, she would think that given real world governments, it is more likely that the use of violence against the innocent will lead to bad consequences, and chose to stop it. This response is somewhat defused by considering a government for which you are told, explicitly, that you know nothing about, in an alien planet far away.

It can be the case that, given more information, it would be right, all things considered, to use force against an innocent. Would you punch a random person in exchange of the future existence of mankind? Most people would.

Initiation of force is prima facie wrong, and it can be right in some cases, but the burden of proof if on those who argue for it.

12.3.1 Sometimes, the use of force has good consequences

Sometimes, yes. But what you would need are occasions where the use of force has good consequences, and it is justified in that circumstance. If successful, that argument wouldn’t justify a government as understood today, but a very narrow governance institution, and that only if the overall institutional framework is better than an alternative.

Suppose there are two kids who will die in a year due to a rare illness. Suppose now there is a person with a peculiar genetic makeup such that he has resistance to the illness. Science says that if experiments are performed on that person, a cure can be produce before one year to save the kids. Here, the consequentialist recommendation is to kidnap this fellow and experiment with him. Everybody else would object to such kidnapping. Consequentialists may want to avoid this by resorting to rule-consequentialism. I have some comments regarding this in the following sections.

12.3.2 Non-consequentialism is arbitrary

Few people follow theories that are fully consequence-independent. Noninterference with other people seems to be one very universal principle, but that doesn’t lead you to consequentialism or utilitarianism.

The claim of arbitrariness could also be thrown at consequentialism: why is harm bad? Why should each of us consider everyone’s interests equally?

In the end, if we understand arbitrariness as “beginning from false or dubious premises”, we would have to focus on ethics, and study the relevant premises. That is outside of the scope of this FAQ. The Principle of Non-Aggresion is a hook to hang pre-existing prejudices

Maybe for some people it is. Indeed, people believe in something like the NAP, and a belief being widely held makes it hard to eradicate if it is false. This claim can also be leveled at political authority.

Some versions of the NAP may require hardcore epicycling to work. But the same thing is done by utilitarians when they resort to rule utilitarianism, or by changing the definition of utility (pleasure, preferences, welfare, etc). In the end, most moral systems clash with some or other strong intuition, this is not a problem only for utilitarianism or the NAP.

That said, full, hardcore exceptionless NAP has its defenders, but not every libertarian agrees with the version of it that the FAQ criticises. Arguments for the NAP are weak

Many are, specifically Rothbard’s and Hoppe’s (Eabrasu 2012). But perhaps the idea that the things that form part of one’s personal projects, body included should not be aggressed against is one plausible enough to be self-evidently accepted (As is the idea that harm is to be avoided). This doesn’t solve the issue of what things become property, to what extent things are property, and if what is property today still is property tomorrow. That’s why you need a legal system for. Libertarians accept the Lockean Proviso, so the situation described for Tristan de Cunha can’t happen.

Here is A. John Simmons, who is no libertarian, but a philosophical anarchist Lockean, interpreting Locke to explain the Proviso:

Fairness, then, seems to generate a conflict. Fair acquisition may result in resources being closed out (or in the requirement that any reminder of them be left common, thereby eliminating rightful appropriation). But fairness also demands a share for later persons. Governments must settle such conflicts. But these conflicts need not be insuperable, since later persons can get their share by getting access to a living. Fairness does not necessarily require the redistribution of initially fair acquisitions. It requires only that persons who cannot appropriate a share are not denied access to their share or room to exercise their rights of self-preservation and self-government. Where these latters goods are precluded, later persons must be allowed to appropriate the property of earlier ones. This conception of a “fair share limit” on appropriation, holding, and transfer seems to me a plausible addition to the mixing argument. (The Lockean Theory of Rights (1994, ch 5))

Here is Nozick

A theory of appropriation incorporating this Lockean proviso will handle correctly the cases (objections to the theory lacking the proviso) where someone appropriates the total supply of something necessary for life (For example, Rashdall’s case of someone who comes upon the only water in the desert several miles ahead of others who also will come to it and appropriates it all.)

A theory which includes this proviso in its principle of justice in acquisition must also contain a more complex principle of justice in transfer. Some reflection of the proviso about appropriation constrains later actions. If my appropriating all of a certain substance violates the Lockean proviso, then so does my appropriating some and purchasing all the rest from others who obtained it without otherwise violating the Lockean proviso. If the proviso excludes someone’s appropriating all the drinkable water in the world, it also excludes his purchasing it all. (More weakly, and messily, it may exclude his charging certain prices for some of his supply.) This proviso (almost?) never will come into effect; the more someone acquires of a scarce substance which others want, the higher the price of the rest will go, and the more difficult it will become for him to acquire it all. […]

The total supply could not be permissibly appropriated by one person at the beginning. His later acquisition of it all does not show that the original appropriation violated the proviso (even by a reverse argument similar to the one above that tried to zip back from Z to A). Rather, it is the combination of the original appropriation plus all the later transfers and actions that violates the Lockean proviso. Each owner’s title to his holding includes the historical shadow of the Lockean proviso on appropriation. This excludes his transferring it into an agglomeration that does violate the Lockean proviso and excludes his using it in a way, in coordination with others or independently of them, so as to violate the proviso by making the situation of others worse than their baseline situation. Once it is known that someone’s ownership runs afoul of the Lockean proviso, there are stringent limits on what he may do with (what it is difficult any longer unreservedly to call) “his property.” Thus a person may not appropriate the only water hole in a desert and charge what he will. Nor may he charge what he will if he possesses one, and unfortunately it happens that all the water holes in the desert dry up, except for his. This unfortunate circumstance, admittedly no fault of his, brings into operation the Lockean proviso and limits his property rights.  Similarly, an owner’s property right in the only island in an area does not allow him to order a castaway from a shipwreck off his island as a trespasser, for this would violate the Lockean proviso. […]

I believe that the free operation of a market system will not actually run afoul of the Lockean proviso. (Recall that crucial to our story in Part I of how a protective agency becomes dominant and a de facto monopoly is the fact that it wields force in situations of conflict, and is not merely in competition, with other agencies. A similar tale cannot be told about other businesses.) If this is correct, the proviso will not play a very important role in the activities of protective agencies and will not provide a significant opportunity for future state action. Indeed, were it not for the effects of previous illegitimate state action, people would not think the possibility of the proviso’s being violated as of more interest than any other logical possibility. (Here I make an empirical historical claim; as does someone who disagrees with this.) This completes our indication of the complication in the entitlement theory introduced by the Lockean proviso. (Anarchy, State, and Utopia)

And here is Eric Mack, discussing the case of the island and other, similar, case.

Here are Brennan and Zwolinski, discussing the proviso. (Brennan even notes that academic libertarians don’t even use the NAP as a basis for libertarianism)

Arguments for private ownership of stuff don’t say that because you have your body, therefore you own it. They explain why you own a body, and then the rest of the things. Bodies are not qualitatively special.

Plus, banning bodily self-ownership as a foundational moral value is question begging. Everyone agrees to some version of self-ownership. The NAP has a weird definition of force, and libertarians tolerate certain initiations of force, and the NAP requires a sausage of exceptions to work.

Definitions are arbitrary. ‘Initiation of force’ is defined as violating rights, and rights come to be according to the libertarian principles of homesteading and contract.

Some libertarians do not tolerate those exceptions, but, admittedly, most do. Problems with NAPping your way through ethics were pointed out decades ago by David Friedman, who was subsequently criticised by Block (2011) (I still agree with Friedman). Rights in theory shouldn’t conflict, if in practice they do, courts will have to try to interpret them the best they can. This is, with different rights, how courts work right now. And consequentialist considerations are allowed to play a role in determining those hard cases.

12.4 Consequentialism is better than the NAP

People are, in general, not implicit or full consequentialists. Not even Scott Alexander. Even self-declared consequentialists do not behave like such, even though it is possible to do so. Consequences matter, and other people matter. But not infinitely.

Also, capitalism is not driven by consequentialism, but by rights-bounded search for entrepreneurial arbitrage opportunities. People are not usually trying to unconstrainedly maximise their (or everyone’s) welfare, even if economists model firms as profit-maximising because it’s close enough to reality. Plus, consequentialism in the way usually depicted in the FAQ is different from ethical egoism, or do what’s best for you, disregard the rest of the world.

12.4.1 You decide what consequences are good

(Consumerism tells you? Huh?) I think there is a confusion here in mistaking economic decisionmaking for ethical theorising. Decisions have opportunity costs, and we face tradeoffs. Those have to, and are, factored in when we think about what’s the right thing to do.

12.4.2 If you care about other people, take them into account in your ethical calculations

Most people do that, but that’s not a purely consequentialistic thing. Plus, this reasoning is impotent vs ethical egoists, or national consequentialists who don’t care about foreigners, or even against family consequentialists. Or… white supremacist consequentialists.

12.4.3 Consequentialism won’t become a war of all vs all because people will agree to a set of rules to avoid so

Yeah, but that undermines previous reasoning earlier in the FAQ. Individually, I’m maximising my good optimally if the rest of the people refrains from maximising their goods and let me maximise mine. What incentive do I have to bind myself? I don’t control other people, so I can calculate the expected utility of each decision, and go for maximising my good anyways. In reality, people would probably coordinate, as I’ve explained above, byt the FAQ is quite critical with overcoming anti-coordination incentives.

12.4.4 That won’t lead to undesirable situations like slavery, decadence or some dystopia

Who knows. Maybe those things do maximise utility. Maybe some people will totally think that the right thing to do, long term-discounted-welfarewise, is for the US to conquer the world, enslave sub-100 IQ people, and coordinate global resources to wirehead humanity in order to get a future of eternal brain stimulation-induced pleasure. (Something that Scott rejects). If such a thing were endorsed by utilitarianism, would you go with that, or accept it as a reductio?

12.4.5 Consequentialism is the gold standard for morality, but maybe trying to implement it, by just doing what we think has good consequences, is not a good idea

Even Scott, the writer of the FAQ, doesn’t really believe this, as pointed out above: he knows that it still misses things. Scott hints here at the act/rule consequentialism distinction. In the next section, I will criticise it.

12.4.6 Read the consequentialism FAQ

I did, and I’m not that impressed.

13. Rights and Heuristics

13.1 Rights are conclusions, not premises. Heuristics. They should only be supported if they bring good consequences

Every argument begins with some premises. Consequentialists have theirs, deontologists have theirs, virtue theorists have theirs, and so on. In consequentialism, rights are (or not) conclusions. In deontology, good consequences are conclusions that follow (or not) from following the right rights structure.

Things like ‘I shouldn’t kill people, except in very particular circumstances’ seem more initially plausible than ‘I ought to maximise aggregate utility’. Now, you may ask for a unifying principle for rights from where to deduce them. No one has worked out a moral system that works perfectly, but libertarians may reply that self-ownership or the NAP or the implications of the fact that you are arguing (Hoppe) are such an axiom. But one can reply by rejecting those axioms, or claiming the reasoning is fallacious, as academic libertarians point out, but the same can be said of consequentialists. See this list for several background philosophies libertarian theorists rely on. Read this and this and this for further info.

So: Right theorists will say that, one of the reason for rights is that rights-respect is constitutive of a good society by itself, separately from consequences. Since a few libertarian theorists are pluralists, they will say that there are a series of goods, and not just “happiness” or “utility”, and that rights are both part of those goods, and means to get them.

13.2 Consequentialism doesn’t imply breaching rights when it seems to smart people that they should be breached, rule consequentialism is the answer.

There are some philosophical reasons  to be against rule consequentialism. If you are smart, and have thought about the issue, and other smart people agree, it seems that it is morally mandatory to violate someone’s rights if that increases utility. If not, you would be tied to respecting the right structure determined at some instant of time, and never revise them.

13.3 Sometimes, you should change which rights to respect: the alchemist example. Rights are warning signs.

This doesn’t solve the problem presented before. How to determine whether a paradigm shift of sorts should induce a reconsideration of whether a right should still be in place or not? How to determine what is such a circumstance in which we should do that? One solution is abandoning consequentialism and directly applying common sense morality: people have rights, and those can be breached if the consequences are really bad. This is no different from the approach most academic libertarians already take.

And pace Scott, a rule against practising alchemy was a bad idea: people pursuing alchemy ended up doing useful things sometimes, leading up to modern chemistry.

13.3.1 Property rights are generally useful, but sometimes they can be breached, okay. But if people did that on their own, consequences would be terrible. Therefore, it is best for a central agent to decide uniform rules for when to violate rights: governments.

If governments are needed to avoid societal breakdown, governments are justified. But is that true? Minarchists say it is, anarchists don’t. Is the welfare state required for that? The welfare state was born about 130 years ago in Germany. The modern welfare state is a post-WWII creature. Before it, it doesn’t seem like society was at the brink of collapse: it was getting better and better. To decide when it is justified to break rules, you need a court system. Many court systems accept extreme good consequences as an excuse.

13.3.2 Choosing when to violate rights is complicated, but we can have some meta-rules for that: at least must be a procedure to get everyone’s input that works. Governments are one such example.

Scott proposes this. Libertarians will reply rule of law: a common law system can play the same role. Governments work well in developed country, not so in many underdeveloped ones. In democracy, there is no will of the people. There are results that are the outcomes of an election process that is designed by people who can vote (not children, not foreigners), and who vote once on who is going to rule, not issue by issue. Unless you’re Switzerland and can pull off large-scale direct democracy.

So democracy fails the universalizability test, unless we restrict the meaning of ‘everyone’ to ‘every adult in a given country’. And still, it can be said that if something seems right to you you should do it. Or if something is wrong, not do it. The rest of the world will react to you, and you knowing this, will take it into account. If there are lots of robberies because people are hungry, people will hire security, and other people may philantrophically provide for them, or even just devise new business models to lower food prices. The system will end up balancing itself.

Regarding government size and well-being: There is no clear correlation between higher taxation rates (or government spending) and standards of living, beyond the fact that only rich societies can sustain a large State.  You can have huge governments like in Kiribati, Cuba, Lesotho or Micronesia and live in poverty, or small ones like in Hong Kong, Switzerland, or Singapore, and live in prosperity. Ideally, we would compare the Nordics to a place with a truly minarchist(<5% GDP) government, and free markets, but we don’t have contemporary examples.

13.4 Governments may make mistakes when acting. But not acting can have worse consequences: the case of the Holocaust and Rwanda.

The Holocaust wouldn’t have happened if everyone had followed a non-interventionist policy from the beginning. No First World War, no Versailles Treaty, no poor beaten Germany, no Hitler, no World War II, and no Holocaust. Pacifism seems to be the way to go.

And most, if not all, great catastrophes resulting from violating rights have been perpetrated by large governments. It’s not that sometimes they violate rights: Some governments regularly do it.

There is an asymmetry in using coercion in war: to use it, it has to be ex-ante crystal clear that it is going to result in the desired consequences, and that rarely it’s the case. And where did all the talk about being careful about not violating heuristics even when it seems it’s the right thing to do?

13.5 Libertarians seem to endorse fighting terrorism, which is expensive, but if instead that money went towards fighting cardiovascular diseases, more lives would be saved. There is no benefit in fighting terrorism instead of CVD. 

I haven’t surveyed libertarians on this, but here there’s one talk by Michael Huemer in which he explicitly makes the point that devoting massive amounts of resources to fighting terrorism is irrational. So I think libertarians should be against that. But I don’t see the similarity with the previous case, unlike the argument being made is just that killing by omission is the same as directly killing, a premise that almost no one accepts, and that would have to be argued for. On what rational grounds is the decision that there is no difference between killing by omission and directly killing being made?

13.6 There’s a case in which libertarian principles are followed and a catastrophe results: there’s a pandemic, and only a guy has the cure, and he charges what he wants for it, and everyone ends up enslaved

On Lockean Proviso grounds, he would have to provide the cure at a fair price to everyone who asked for it, and it would be justified to coerce him to do so.

This, however, is a very good example of why ‘hard’ libertarianism is problematic. Regular varieties of it are unaffected by this issue.

13.6.1 Okay, the story is not realistic, but it is still a reductio ad absurdum. The fact that the heuristic “never initiate force” works is only limited to some cases. The circumstances in which libertarianism works are contingent

It’s a good reduction. But it doesn’t work against academic libertarianism. If the world wasn’t the way it is, then we would factor that in into the conditions required to violate rights.

13.7 Being poor in countries without government welfare sucks. Private charity is not solving third-world poverty. Private charity can’t plausibly increase as much as required, as some easy math shows.

Most developed countries do have some kind of government welfare, so it’s hard or impossible to find the poor Scott is looking for. There are still homeless and really poor poor people even in advanced social democracies like Denmark, as mentioned above. (Poor after taxes and transfers, that is)

The calculations Scott suggests were made for Spain in a book here that proposed a detailed transition to a minarchist state (5% GDP) in 50 years, covering specific proposals and budgets for education, healthcare, pensions, etc.

The math itself is right, but one premise is wrong: spending on welfare is not really helping the poor. In the US, welfare for the poor is the EITC, TANF, SSI,food stamps, housing vouchers, and a child tax credit. In total, $212bn. So the increase in money in the hands of the private sector, would be able to offset the increase in charity required.

Maybe more refinement could be done: The precise calculations outlined by Scott were done in the previously cited book for Spain, which has a higher market-poverty rate than the US. And Switzerland, which I’ve been citing as an example all along, as a market-poverty rate (by relative standards, I couldn’t find pre-market absolute poverty rates) of 15%, even lower than US’s post-tax and transfer.

13.8 People who make bad decisions don’t deserve to suffer

Agree. But insulating people from the consequences or their actions is incentivewise problematic.

Part E: Practical Issues

The Argument: Allowing some power to government won’t lead us to tyranny. Libertarians should instead work within the system to oppose regulations that don’t work, and accept regulations that do work.

The Counterargument: This argument is made not by minarchists, but anarchists. The problem is not tyranny, but a government larger than it should be. In advanced countries, it would be rare that the government became tyrannical. Instead, there is at least one historical example of a country that began on the premise of limited government, and ended up with quite a larger government: the US. There are reforms that can be made to enhance the ways democracies work, but instead of going for a second-best, libertarians will still go for the first-best. Everyone else should follow.

14. Slippery Slopes

14.1 Giving some power to government won’t convert it into a tyranny in the long run

As explained above, the problem is not tyranny, but overreach ( if some government functions are accepted).

14.1.1 If we let government have its way, it won’t engineer a collapse into dictatorship

I don’t know if there are many libertarians endorsing this. I agree with Scott’s view of politics.

15. Strategic Activism

15.1 If you are smart enough to be a libertarian, you should use your intelligence and energy to enact a proper system

Given that votes have little power, people have no incentive to do that. This is a huge coordination failure – of the sort that are used to criticise libertarianism. It’s not an utter failure, though, for the same reason the failures pointed out in markets aren’t that bad.

Instead, smart non-libertarians can work to help make the transition to libertarianism :_)

15.2 You can’t say government cannot be improved, there are people proposing clever arrangements for government

Futarchy is replacing one piece of government: policymaking. Governments are still governments made of people, and will still have problems. My point isn’t that this wouldn’t be better. Surely it’s possible to improve government, and futarchy is an interesting idea that I would like to see tried, but it will still be a monopoly on force with political authority, and libertarianism may still dominate it: the market process generates information by itself. If not for this, it would be possible to improve upon capitalism by having a futarchic central planner committee.

Nonetheless, the fact remains that the quality of government in a few countries doesn’t seem to be improving: the rate of approval of US Congress is at a historical minimum, and Trump has shown what happens when a democratic system works as designed. Perhaps it’s time to start advocating epistocracy, instead of democracy?

And altough this issue is orthogonal to this disquisition and raising this borders on trolling, the nifty theorems of economics mentioned are not necessarily that true: Tetlock’s forecasters do better than prediction markets, altough prediction markets remain really good. (Atanasov, et al. 2015)

15.3 Improving government is hard, but libertarianism is politically impossible

Incremental attempt to reduce government are also more viable than transitioning tomorrow to libertarianism. The problem is not just that reforming the government is difficult. Some libertarians do have reforms for making governments better, even if those libertarians don’t want any government at all as a first best. The real problem is that the best possible government is still no match for the best possible libertarianism, and then if a society is able to overcome the many coordination problems that surround politics in order to get good government. Why can’t it also solve the coordination problems you wanted to set up government for in the first place?

15.4 Drawing a line in the sand and banning all government intervention doesn’t work. If libertarians want to have some impact, they should take a case by case approach to policies. People won’t listen to people who are against every single policy.

Some libertarians say this: the US government could be taken as a refutation of the theory of limited government. But the conclusion is then to go for anarchism.

As for the later question, I quote Huemer:

Why is the proposal of this chapter not similarly utopian? Why is it more realistic to expect that citizens convinced of the illegitimacy of government will work to abolish their government than it is to expect that citizens apprised of the flawed policies implemented by a democratic government will work to perfect their government’s policies?

The answer is that acquiring awareness of the illegitimacy of government in general is much, much less cognitively demanding than acquiring sufficient awareness of the specific policy errors of a particular government to enable one to make rational plans to correct most of those errors. To realize that government is illegitimate, it suffices to accept the arguments in this book. But to identify most of the specific policy errors of one’s government would require detailed familiarity with thousands of statutes and regulations; dozens of government agencies, boards, and commissions; and hundreds of political figures. One would have to update this knowledge continuously throughout one’s life to take account of each new action of each arm of the government. It is much more realistic to hope that a consensus could be reached on a single philosophical principle, the rejection of authority, than to hope that a consensus could be reached on the specific flaws of most particular government policies.

16. Miscellaneous and Metal

16.1 I disagree with you. How should I engage with non-non-libertarian arguments in a way that is most likely to change your mind?

Being nice, in general. Read academic libertarian philosophers, and read about the history of how things worked before States began doing them. Don’t assume consequentialism entails a rejection of libertarianism without having fully explored the mechanisms a libertarian society has to solve the many problems that may arise. And furthermore, don’t assume consequentialism to argue for a political philosophy, it will make your job harder. Begin from shared  moral premises, and work out an equilibrium between principles and intuitions. As Scott says, you would have to prove your standards are right from first principles. So far, no one has succeeded deriving a full moral system from scratch. In the future, I might therefore do a critique of the non-consequentialist FAQ. Alternatively, you can show that a realistic libertarian society will be really bad compared to a social democracy, or an alternative system. I think that should convince people to abandon libertarianism.

Don’t assume smart libertarians are the cartoon libertarians you find angrily commenting in internet posts or twitter. Like with every ideological group, there are tiers of sophistication.

16.2 Where can I go to see a rebuttal to this FAQ?

So far, there aren’t

16.3 Where can I go to find more non-non-libertarian information?

Well, non-non-libertarianism is about answering bad objections to libertarianism, so that really depends on the arguments. For good libertarian arguments, read books. Specifically, Mike Huemer’s The Problem of Political Authority, Jason Brennan’s Why not Capitalism? and Libertarianism: What you need to know, Brennan and Schmidtz’s A Brief History of Liberty, G.H. Smith’s The System of Liberty, Loren Lomasky’s Persons, Reasons, and the Moral Community, Nozick’s Anarchy State and Utopia, and D. Friedman’s The Machinery of Liberty.

For a survey of academic libertarianism and classical liberalism more broadly, see Mack & Gaus (2004 , ch. 9), Zwolinski & Tomasi (2014), and Brennan & Tomasi (2011).

Also, I address this to libertarians: Libertarians should be aware that the best libertarian theorists are not Rothbard, Hoppe and Rand. They are not very good, actually. Some authors you should look to are: Nozick, Mack, Lomasky, Huemer, Brennan, Zwolinski, D. Friedman and if you can read Spanish, Rallo.

Conversely, libertarians should try to read or be acquainted with the steelmanned versions of the anti or non libertarian arguments. This means reading Rawls, G.A. Cohen, Jeffrey Friedman, Samuel Freeman, the Non-Libertarian FAQ, and Peter Singer. Or even Pettit and Kymlicka. Of course, they also have to read papers to grasp the empirical details of how things actually work.

In general, if you think governments can make things better, read some history and see long-run historical trends, and how life was before the welfare-regulatory state existed. The systems we have today are not the result of careful empirical work and institutional design, but of war, emotional politics, and institutional inertia. It is naive to assume that the contemporary Nation State is the better and ultimate form of social organisation to exist.

Consequentialists should be willing to explore alternatives, including minimal states, city-states, polycentric orders, and anarchy, instead of focusing that much in within-the-system policymaking.

16.4 How can I respond to this FAQ in some way?

Send me an email to reverse(jritra) at gmail dot com.

16.5 And what about the Metal

Here, have some.


I hope you have enjoyed this FAQ. I think both libertarians (especially the sort of libertarian Scott has in mind) and non-libertarians will be able to learn something from it.

I don’t actually expect you to become a libertarian by reading this. I have not argued for libertarianism here, just explained that the arguments put forward by Scott do not debunk libertarianism. For example, I don’t expect you to believe right now that private healthcare is better than public healthcare. I haven’t discussed the arguments economists put forward in that debate, or done a full literature review. But I hope to have convinced you that it may be workable.

The way you should think of libertarianism right now should be between two extremes: one, full acceptance. And two, seeing it as an interesting idea that maybe should be tried in a tiny place far away. This is Scott’s current position here. Defending libertarianism is usually done by historical examples, or picking examples here and there, as there is no country that is currently libertarian. Hence, debates about its viability or desirability tend to drag on forever, as there is no hard, conclusive evidence on the matter, like social democrats have with the Nordics.

I want to finish by addressing one thing Scott says in the previously cited post:

And this leads me into one of my deepest problems with libertarianism and anarcho-capitalism: why should it work?

I don’t mean the sort of “why should it work” where you answer with specific reasons why no, monopolies won’t form, and no, people won’t routinely sell themselves into slavery, and no, protection agencies won’t form a new feudal ruling class, and no, people won’t bash their heads against public goods problems and externalities forever without any market solutions appearing, and no, the poor won’t starve to death. I mean the very Outside View question of “why is it that, by coincidence, not using force is an effective way to solve all problems?”

Good governance is a really really hard problem. The idea that the solution to this problem contains zero bits of information, that it just solves itself if you leave people alone, seems astonishing. Even if we agree that capitalism works very well by incentivizing companies to do what the consumers wants, there are still a lot of peripheral issues which that just doesn’t cover. Friedman for example is a strong supporter of child rights, because children should mostly be free from coercion from their parents, and that children treated this way turn out better. Now in addition to solving governance with zero bits of information, you have solved optimal child-rearing with zero bits of information. That is implausibly impressive.

Given that the universe is allowed to throw whatever problems it wants at us, and that it has so far gleefully taken advantage of that right to come up with a whole host of very diverse and interesting ones, why is it that none of these problems are best addressed by a centralized entity with a monopoly on force? That seems like a pretty basic structure from a game-theoretic perspective, and you’re telling me it just never works in the real world? Shouldn’t there be at least one or two things where a government, or any form of coercive structure at all, is just the right answer? And can’t we just have a small government that does that?

This is a very good question. Some months before writing that review, Scott wrote his now famous Meditations on Moloch. He sees coordination problems everywhere, and not even governments can fix all them: they are also subject to perverse incentives. The solution, then, is an external unmoved mover, an artificial god, to ensure people do cooperate when they want to cooperate but don’t have the incentives for it. And since that is really hard, the whole piece has a tone of existential angst, of fear and trembling at the idea of unsolvable coordination problems, not evil, as our greatest problem.

There may be an answer to that. But before considering it, think of the socialist calculation problem. Imagine we’re back some decades, and someone writes this to us:

Yes, markets are very good. But hell, how on Earth are they so good that central planning does not work in any industry? Not bread, not steel, coal or oil, not widgets, cars or aircraft. Why? Central planning has advantages, like economies of scale, solving some coordination failures, and reducing duplicities. Why is is that for every conceivable industry you are saying the market works better? And why capitalistic management? Why not worker ownership? How do you know capitalistic ownership works best in every single case?

And I say: If central planning is efficient, that will appear in the market. If worker cooperatives are efficient, that will appear on the market. If worker cooperatives are efficient, that will appear on the market. The market dominates every single alternative, because the market structure will transform into the alternative if the alternative is better, and then readapt if necessary. Today, you might have two big oil companies. Tomorrow, you might have three or eight. The day after tomorrow, maybe none and we’re all using solar energy. At the same time, you can have worker owned firms, anarchosyndicalist firms (like Valve), decentralised firms, with extense supply chains, or firms that do everything inhouse (like SpaceX).

This is basically the First Theorem of Welfare Economics. But then, it’s possible to say: yeah, but that holds if there is perfect information, perfect competition, and no transaction costs.

It is possible to make two counter arguments here. One, of comparative institutions: Governments are not benevolent social planners. According to the Soylent Green Principle of Institutional Analysis, governments are people. It’s possible that the optimal institutional arrangement is one in which markets failures are allowed to exist, but governments failures aren’t, by limiting the role of the State.

And the second, more interesting one, is that market failures are profit opportunities for entrepreneurs to solve. It’s not that there is a series of conditions that markets need to work properly, it is that an institutional framework that allows markets to work will lead to those conditions: enough competition, enough information, and sufficiently low transaction costs. People are able to step outside the games they are playing, rearrange the incentives, get into the games again, and cooperate.

If the argument for laissez-faire conditions was a consequence of behavioral assumptions of super heroic behavioral assumptions like you are a fully informed and perfectly rational chooser that exists in a frictionless environment, which is how a lot of people think the argument goes, then all I need to do is undermine the rationality of agents and also the function of the price system as being a perfect guide, so the perfectly competitive model.

But from Adam Smith, all the way up to Vernon Smith, that’s never the way that they couch their argument for a relatively free economy versus an interventionist economy. They always couch it in terms of comparative institutional analysis and if you look at Adam Smith and the various different examples that he gives, it’s never the case that he relies on a perfectly-informed actor operating in a perfectly competitive market yields an argument for laissez-faire. (Boettke)

Most people see markets as one mechanism and government as another mechanism, and you have to see which one works best in any case. I think a better way of thinking is thinking of markets as a meta-mechanism, that enable many sub-mechanisms (including government-like institutions) to arise from the bottom-up (Pennington, 2013). Another meta-mechanism is to have everything within a government, which is what we have now, altough that, in turn, is nested inside a state of international anarchy.

The solution to the “Who will watch the watchmen?” problem is not to designate special watchmen, but have everyone, within the domain known by that someone, be a watchmen. This, especially when coupled with some moral norms (people are not and ought not be fully selfish) is what makes libertarianism the best possible, theoretically and practically, basis for social order and cooperation.

As you may see in this FAQ, most of the disagreements between libertarians and the rest are empirical. It should be possible to study, and agree on how each system would work. Perhaps it’s the case that overall libertarianism is better except for the poorest 5% of society, who are somewhat worse off than in a social democracy. Or perhaps it’s also better for them. After that is done, we can resume arguing which values are correct.

Some of my answers may have felt like cheating. Am I really giving libertarian answers to the problems Scott raises? Am I a master handwaver?

I have  rejected some absolutist interpretations of the NAP, I have accepted the need for some regulations that could be market provided, and I’ve praised the Nordics above the US model in some aspects, and so on. I claim that most of the critiques, even if successful, can be addressed by a minimal state.

I think the reasons and explanations offered above are perfectly compatible with the kind of libertarianism proposed by academic libertarians, and so while there are serious problems for the sort of cartoon libertarians that may be the target of the FAQ, the arguments presented there fail to refute more rational and informed variants of libertarianism, while at the same time casting some doubt on its alternatives.

This entry was posted in Blog. Bookmark the permalink.