How can economics explain how security can be provided?
For “Ask an Economist” this week, I have a question from Luis in Mexico City. He says:
I’m an assiduous reader of the FEE regular material. I write to you from Mexico City regarding your very liked column: “Ask an Economist.”
For a long time my head has been turning around trying to understand the puzzle of how are economic science and security related (in the most generic sense of this last term)?
So let’s dive into the relationship between economics and security. Since Luis asked about the most generic sense of the term, I’m going to assume that he means a general protection of life and property, since that’s the most generic way to describe security. To do so, we’re going to have to start with the simple, textbook view, then we’re going to move into a more nuanced view of property rights and security economics.
The Simple View: Security as a Public Good
Let’s consider the textbook view. First, security, as a good, is generally considered to have two key attributes.
First of all, security is the kind of good where your use of it does not necessarily take away from someone else’s use of it. For example, if a security guard monitors your neighborhood, his protection of your house does not diminish his protection of your neighbor’s house. If anything, the two actions are seen as complements. By protecting your house from a criminal, he also makes your neighbor’s house more secure.
Economists call this attribute non-rivalrousness in consumption. Your consumption of the good does not harm another person’s consumption. Enjoying a street light is non-rivalrous. If you use a street light to read, that does not prevent someone else from doing so too. Sandwiches, in contrast, are rivalrous. If you eat it, no one else can enjoy it.
Second, security tends to lump in lots of people, whether they like it or not. For example, imagine the US is protecting citizens from invasion. Now let’s say you decided not to pay your taxes for the year. Since you chose not to pay, can the US decide not to protect you? Well, it’s not like the US can announce to enemies abroad, “Here are the houses we’ll allow you to destroy because they didn’t pay taxes.” Doing this would be a security risk for the whole country, so it isn’t on the table.
Economists call this attribute non-excludability. You typically cannot exclude people from the benefits of national security. Oftentimes radio is another example of a non-excludable good. It’s very difficult to prevent people from receiving radio signals so long as they know how to use relevant equipment. Internet service, on the other hand, is excludable. You can keep someone from accessing the Internet unless he pays.
Goods that are nonrivalrous and nonexcludable have been described by professional economists as “public goods.” The name doesn’t imply that the government does provide them. It’s more a commentary on the fact that the goods in question will be broadly consumable without limit if they are provided.
Since the consumption of these goods is too costly to prohibit selectively, there will be an incentive for people to free-ride off the purchases of others. The concern is, if everyone is trying to free-ride, no one will be willing to foot the bill and the service won’t be provided.
What makes this a big deal is that security is necessary for productive economic activity to take place. Without secure property rights, individuals cannot credibly commit to exchanging goods and services (because exchange implies that you have property of the thing in the first place). So, security is necessary for exchange.
Consequently, textbooks view security as the ideal example of a public good which won’t be provided at the proper level absent government intervention.
A More Nuanced View
Oftentimes, textbook views of things are overly simple. When it comes to security, I believe that’s the case. Work by George Mason University economist Christopher Coyne gives us good reason to be skeptical that security should be viewed as a public good with government provision as the only solution. In his paper Lobotomizing the Defense Brain, Coyne makes this compelling case.
How so? Well, first of all, Coyne argues that whether or not security is excludable is somewhat definition-dependent. If you consider national defense as a single good, it’s probably true that no one can be completely excluded from it. However, security is not just one homogeneous good. It’s composed of many small goods which are neither non-rivalrous nor non-excludable. He says:
If one changes the context slightly, a different outcome emerges. For example, each individual anti-ballistic missile is rivalrous and excludable (Cowen 1985: 56, Hummel 1990, Hummel and Lavoie 1994: 355). It is rivalrous because the same missile cannot protect two geographic areas … Further, the missile is at least partially excludable because one can, in principle, protect paying states, cities, or localities while excluding non-paying locals.
If security is, even in part, rival and excludable, the standard public good story does not apply.
Another issue with the idea that the government should provide the good of security is that it assumes that the version of security the government provides is a good. Can we say that the misadventures in Iraq and Afghanistan provided more of a “good” than a “bad” to the typical American in terms of security? I’ll leave you to decide.
Coyne also points out that even if security were a public good and even if the government could provide it as a good (rather than a bad), the government still has the issue of determining the right quantity of the good to provide.
We have all sorts of goods that people like in our economy: food, healthcare, education, entertainment—the list goes on. However, we cannot produce unlimited quantities of these goods because resources are scarce. Therefore, it’s possible both to overproduce and underproduce goods.
Because of the nature of the economic calculation problem, the government lacks the ability to determine what amount of security would best satisfy consumer preferences. The economic calculation problem in security makes the case for public provision much weaker.
Can Private Markets Provide Security?
Markets need secure property rights, but can secure property rights be established without governments? At the very least, it seems like they can in certain circumstances. Most of the security you encounter in your daily life is private. From home security systems, to locks, to private security guards at malls and events, private security solutions abound.
Some argue that these things only exist in the shadow of the government itself, and therefore there is no reason to expect that they would exist absent the government backing them. For example, private security officers may only be seen as credible because the government would uphold your right to protect property with security in court.
However, we do have some evidence that even without government presence, private security can arise. Terry Anderson and PJ Hill argue in their paper The Not So Wild, Wild West that western-moving homesteaders in the US frequently set up successful, voluntary peace-keeping institutions without any reasonable expectation of the US government having the resources to support them. This paper represents just one example of how private provision of security can work.
So, to Luis’s question, the relationship between economics and security is a complex one. The simple logic of public goods and security implies it needs to be provided by a state to be successful because of the nature of public goods.
However, when we let the analysis become more complicated, we see that the theoretical case that security is always and everywhere a public good is weaker than it first appears, and there is significant reason to doubt the government’s ability to allocate scarce resources efficiently.
Whether that means a free market approach to security would work better is a matter of active debate. What seems clear, at the very least, is that there’s probably a better solution to the security problem than the status quo.
