Thursday, July 7, 2011

Russ Roberts on "A Theory of Government"

He writes:

"You can think of two theories of government. One theory is that government exists to correct externalities and provide public goods. The other is that government uses the language of helping people to justify giving stuff to the politically powerful out of the pockets of the rest of us. Here’s some evidence for the second theory."

And provides this video:

I write this in the comment section:

"Those options seem somewhat lacking.

How about government is one of many emergent human institutions that evolve over time to solve externalities and public goods problems, but which also provide the opportunity for rent seeking behavior.

This, I would argue, has the most evidence.

1. If you look generally at what governments do, public goods and externalities are heavily represented.
2. The evidence for rent seeking and predation doesn’t need to be recounted here – it’s obvious.
3. The degree to which governments provide public goods vs. work as an opportunity for rent seeking is very closely tied to institutional structure. Institutional structure is at least somewhat evolutionary: fit institutions that provid institutional resistance to predatory behavior and rent seeking seem to survive in the long run.
4. The benefits of rent seeking are concentrated and the costs are diffuse. The benefits of solving externalities are diffuse and the costs are concentrated. So institutions are unstable. Cultures with norms and culture that enable collective action have an advantage in developing stable governing institutions.

I think I win.

I’d stack that theory of government against a libertarian political piece about taxis any day of the week.

Let’s stop this ridiculous charade of identifying two elements of government, giving them the label “two theories of government” and then pretending that both elements aren’t strongly supported by the body of social science and an enormous swath of social scientists


  1. (1) A lot of public goods and externalities are claimed yes ... the evidence as to whether are or not is another thing entirely. True public goods and externalities are far more rare than is claimed.

    Once that premise falls apart your entire argument falls apart.

  2. What Gary said! You provide no evidence to prove that government is largely about externalities and public goods. Again, you provide no evidence that it does a better job at this task than would markets and private property. Your argument shatters into a million pieces because I think reality is just the opposite of your assertion. I think I win.

  3. If you're this confident now, I'm afraid to see what your blog posts will be like when you're 50 and tenured, Daniel...

  4. It seems you're tilting at windmills, Daniel. Roberts is not indicating that government only exists for purposes of theory #2. Rather, that the allocation of taxi-cab medallions falls under #2, when it is directed at a #1 problem.

    But, yes, sometimes even the most obvious lessons are well to be repeated. I mean, I think students get rather bored of hearing about price controls in the 1970s or rent control in NYC.

  5. said -
    So, based on the text of the post itself my understanding was that Russ was presenting two theories of government, not two elements of one theory. This was my impression because those were the words he used.

    If he meant what I was saying, then that's great - clearly we agree, and I love it when two people coming from fairly different perspectives find something they can agree on.

  6. Bob -

    re: "and tenured"

    Alas, now you're appearing more confident than I'm allowing myself to be. I'm excited about my career, but AU is no NYU on the resume.

  7. Daniel,

    I don't mean to get into a rhetorical analysis of Roberts' blog post, but I think it is telling that Roberts says "government exists" in theory #1 and "government uses" in theory #2.

    I am not sure how #2 could be a theory "of" (of meaning why it was formed) government, other than a theory of how government is used by rent-seeking actors.

  8. As history has showhn, favoured businesses and associations use government regulatory bodies to maintain their cartels and stifle competition resulting in a net deadweight loss for society.

    I think I win.

  9. Anonymous -
    That evidence seems to support my theory as well as yours, so that particular piece of evidence strikes me as a draw.

    Can your theory explain why government produces national security and education but not furniture and apple sauce?

    Presumably buying off the population (as Gary has suggested) is as easy with one as it is with the other.

  10. Can your theory explain why government produces national security and education but not furniture and apple sauce?

    I would think it is for the purpose of creating fear and extracting obedience from the people. They don't do apple sauce and furniture for the same reason a mafia doesn't do it.

    I am part serious and part joking, but most of what government does in the name of education and national security is blatant overreach. It is unnecessary, and it is only done for the reasons stated in my first paragraph.


All anonymous comments will be deleted. Consistent pseudonyms are fine.