Why I’m a Libertarian Part 3/4: Negative Externalities of Government

For part 1 of this series, see The Coercive Nature of Law.
For part 2, see The Efficiency of Free Market Capitalism.

The most popular argument against free markets among economically literate people is market failure. Non-libertarians who understand economics generally concede that free market capitalism is efficient under ideal circumstances, but argue that in many real-world markets, these conditions do not hold. Therefore they favour markets in principle, but want the government to step in to correct these market failures.

By far the most important so-called market failure is the existence of externalities. In my previous post, I explained why free markets are efficient if property rights are drawn such that any use of property and any transaction does  not damage or improve the property of third parties. When this is not the case, we have an externality. There are both positive and negative externalities, and both can lead to sub-optimal outcomes, as the following examples will illustrate.

Suppose you live in pure free market society and there is a contagious disease against which you can protect yourself through vaccination. Getting vaccinated costs $150. Based on the severity of the disease and the likelihood of contracting it, you value the benefit of being vaccinated at $100. However, because the disease is contagious, your getting vaccinated would also reduce the risk of further spreading the disease. The summed benefit to everyone else from removing the risk of getting infected by you is another $100. This means that the total societal benefit is $200, while the cost is $150. However, since this second $100 is an external benefit (i.e. it accrues to third parties who have no say on whether or not you get vaccinated), which you presumably don’t take into account in your decision making, you choose to forgo the vaccination. Thus, free markets can lead to a sub-optimal outcome in the presence of positive externalities.

Next, consider a scenario where you’re running a factory. You calcuate that producing an additional 1,000 widgets per month will cost an extra $80,000 and bring in an extra $100,000 in revenue. However, increasing your production would also increase the air pollution you produce, which will slightly reduce the quality of life for people who live near your factory. There are 100,000 people living near your factory and each values the reduction in quality of life as equivalent to losing 50¢ per month. This is an external cost, which presumably doesn’t alter your resolution to increase widget production. From your own perspective it is profitable to do so, but from a societal perspective it is inefficient since the total cost of $80,000+100,000*$0.5=$130,000 exceeds your extra revenue. This is a case of negative externalities leading to sub-optimal outcomes.

It is possible for government to improve on this. In the first example, government could subsidise the vaccination by more than $50 to make it worth your while to get vaccinated. In the second example, they could tax your extra pollution at more than $20,000 to prevent you from increasing your production. The ideal level of tax or subsidy would be equal to the size of the externality. This idea is called Pigovian taxation (or subsidisation). Such a tax or subsidy will cause decision makers to take into account external costs and benefits. Perfect Pigovian taxes and subsidies would eliminate the problem of externalities.

The big problem in practice is that, unlike in the examples provided above, you don’t know the size of the externality. How would you go about determining how much someone’s quality of life drops from air pollution, and how do you assign a monetary value to that? The best you can do in the real world is make some very rough estimates. And getting these estimates at least roughly correct is essential, because otherwise you might end up making an externality worse instead of better. You also have to keep in mind that there are costs associated with taxation: Not only to you have to pay government workers to oversee and adminster the collection of taxes, you also impose costs on individuals and firms for complying with taxes and you create incentives for tax avoidance and evasion.

There are further problems with the above examples, which are left as an exercise for the reader. The point is that the existence of externalities merely shows that free markets could lead to sub-optimal outcomes, and that government could improve on the market outcome. It shows that perfect government can improve on imperfect markets, but it doesn’t show that real world governments are likely to improve real world markets.


In the market, externalities are the exception, rather than the norm; most benefits and costs are internal. In government and politics, externalities are the norm.

Suppose you are the president of country with a population 100 million people. A law is proposed which will benefit some special interest group to the tune of $10 million per year, while imposing a cost of 50¢ p.a. on every citizen. A net loss of $40 million to the nation, but since ordinary citizens can’t be bothered to oppose a law with such minuscule effects on them (and probably most of them won’t ever become aware of the very existence of the law), it won’t affect their voting behaviour. The special interest group, on the other hand, offers you one million in campaign contributions for your re-election if you throw your support behind the law. What do you do? Hint: 1,000,000>5.

The above example illustrates the externalities faced by politicians. Unlike in the market, in politics almost all the benefits and costs are external, since by its very nature, politics revolves around deciding about other people’s lives. I believe that politicians tend to be considerably less moral than average, but even if we assume that this is not the case, politicians will still be guided more by personal gain than interest in the common weal. Their decisions will be determined by what keeps them in the good graces of voters and powerful interest groups. Sometimes what is good for a politician’s career coincides with what is good for the country, but other times it does not.

Nor is there any strong connection between which policies are beneficial to the country and which are beneficial to a political career. Well-informed and rational voters might provide the right incentives to politicians, but the vast majority of voters are neither well-informed, nor rational about politics.

This is not an accident. When you buy a car, you have an incentive to inform yourself about the costs and features of various models since you will get exactly the car you choose. Both the costs and the benefits are internal. (There are some externalities, e.g. with pollution and road safety, but these externalities are quite small compared to the internal costs and benefits.) When you vote, you do not have a strong incentive to inform yourself about the policies of the various candidates and parties since who you vote for is extremely unlikely to change who gets elected. The costs of getting informed are internal, but the vast majority of the benefit is external. This is why economists say that voters are rationally ignorant.

Even worse, voters are also irrational. Insofar as people do get informed about politics, it is mostly because they enjoy it in a fashion similar to a spectator sport. They want to cheer on their favourite team, and thinking rationally about the shortcomings of their candidate or party would sap away most of the enjoyment. In his 2008 book The Myth of the Rational Voter, economist Bryan Caplan lays out his case for why voters are “rationally irrational” and identifies areas where voters are biased. This means that politicians, in following the will of the people, end up systematically implementing bad policies.


Perfection is denied us on this side of the Garden of Eden. While it is true that markets are imperfect and sometimes lead to suboptimal outcomes, we have good reasons to believe that markets are pretty good in most cases. The situation for government is the reverse. Not only do we not have any good reason for why government involvement should improve things, we have good reasons to believe that government usually mucks things up. Markets are not perfect, but they are the best we have. Combined with the moral arguments presented in Part 1, we have a very strong case for limiting government involvement as far as possible. In the next and final part of this series, I want to talk about the beautiful and harmonious society we can achieve if we give freedom a chance.

(Some of the links above are affiliate links. If you click on them and decide to buy a product, I receive a small commission, at no additional cost to you.)

This entry was posted in Libertarianism and tagged , , , , , . Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *