Saturday, May 28, 2011

Friedman on learning from experience and rights

David Friedman had a good post last week on a commencement speaker that he thought failed to learn from experience. Essentially, the speaker felt that everyone had a legal right to "health care, education, and financial aid", and had personally had trouble dealing with (presumably public) bureaucracy getting these sorts of benefits for his disabled adopted daughter. Friedman suggested that he didn't seem to learn from his experiences when he continued to call for public provision of these sorts of things.

Although I take his point and never promote "positive rights" thinking here, I think there are a lot of problems with this sort of logic. First and foremost, if you are the sort of person that thinks people are entitled to these sorts of things - for whatever reason - you're not failing to learn from experience if you critique a public system that fails. Let's say you think these things should be provided by the market instead. If the market leaves a lot of people unprovided for, and frustrates the people it does provide for with red tape, is that "empirical evidence" against your anti-positive rights prespective? Of course not. You console yourself with arguments that the market will correct itself as people react to these failures. That's really all the commencement speaker is doing here. He doesn't seem to me to be failing to integrate his experiences - he seems to be considering his experiences deeply and responding to them. People aren't bound to develop their understanding of rights from empirical experience with the efficacy of certian systems.

The other thing that bothers me about this sort of logic is that it presumes that whenever a government does something, people always generalize that to public provision in general. This is really an odd way of looking at things. When a particular firm fails to satisfy you, it's not sensible to say "well I guess the market can't provide X". Instead, you seek out firms that do the job better and console yourself with arguments that the price mechanism will privelege those firms. That seems to me to be all people are doing when they see a failure of government but still want to seek out a better public solution. It would be awfully myopic to take one failed approach to public provision and rule out that option. Kate and I are dealing with really obnoxious extra charges that our phone company adds no matter how many times we call them and remedy the problem. If it gets bad enough, I may consider switching phone companies - but I'm not going to start demanding state-run phone companies! Similarly, when the government fails at something that I consider to be a valid public service, my first reaction is going to be to try to change how the government works, not abandon my understanding of the appropriate role of government.


There are a few other things I don't think people consider enough in these sorts of discussions:

1. There's a difference between "positive rights" and "legal rights". It may be reasonable to say that "everyone has a right to a good education" and understand that that right will go unsatisfied sometimes, without arguing that the government ought to provide that education. Anarchists, after all, generally acknowledge some set of "rights" without expecting any government provision.

2. Assessing the value of public services needs to be done with (1.) marginalism and (2.) general equilibrium effects in mind. I cannot stress this enough. It's meaningless, for example, to show that private school children do better than public school children. Of course they do. If they didn't do considerably better they wouldn't shell out tuition for it. Even more cost effective private schooling shouldn't compare averages with public schools. The appropriate comparison isn't between an existing public school and an existing private school. It's between a system of public schools (presumably with some private) and a system without public schools. Too many people who talk about this subject take a marginal effect on the intensive margin (putting a given similarly placed student into private or public school) and use it to try to say something about the extensive margin (comparing a system of public or no public schools). These are entirely different research questions. You can have too much public school on the intensive margin but still have the right approach (public provision) on the extensive margin. These intensive/extensive margin differences matter tremendously - they lead people to say dumb things like "recessions are caused by high wages".

3. Public and private, and government and market are not mutually exclusive. You can have market-based public intervention (i.e. - charter schools, carbon taxes). You can have the coexistence of public and private provision (NASA and SpaceX). Advocates of certain public solutions should not be confused with opponents of private solutions.


  1. Since public life and private life are not mutually exclusive, we must ask why people adopt two different standards for both.

    Why does Keith Obermann complain about income inequality, when he is paid $30 million a year? Why does he not forfeit 90% of his salary or give it away to charity or give it to some random household? Same for any major public intellectual who complains about the same.

    Why do some scientists talk so much about the need to go to Mars, but never foot out their own bills, take out their own funds and resources, take a mortgage on their own house, go out and request donations,.etc and actually start their own space program, instead of coming on television and saying, "We NEED to go to Mars. SOMEBODY start a Mars mission. Anyone but us!"

    I don't have many kind words for Bollywood actors and filmmakers, due to their incredible lack of talent, but these people pay huge figures in advance tax payments. Privately, they compete with each other to see who is willing to give the most to the government. And yet, many of them never make any political statements, never come on television to demand "more taxes for the rich" or whatever. They just pay more taxes themselves quietly, and move on. The media unearthed this little detail only recently.

  2. Well one major reason is the whole reason for public institutions in the first place (as well as the development of markets, for that matter): collective action problems. Neither Olbermann or the scientists could fix inequality or get to Mars on their own. Now, if there were some moral imperative to do these sorts of things, that shouldn't matter. But if it's goal-oriented rather than moral-motivated it does matter, and things won't get done if the goal can't be feasibly achieved individually.

  3. Well look, Larry and Sergei of Google didn't go out to the public and keep saying, "We need a means to allow people to have greater ease in finding what they want on the internet".

    No, they refused a well paying job and future career, put their own money on line, pitched it to several venture capitalists, raised funds from them, and did it.

    Even if it is goal oriented and involving things on a large scale, why can't it be done on pure initiative?

  4. The social objective of skilled investment should be to defeat the dark forces of time and ignorance which envelop our future.

  5. Just want to make a quick point in support of your point 2) under the fold...

    The direct comparison between government and private schools is flawed because of selection bias and identification problems. Crude analogy: You don't say that handing out glasses to every child will improve their marks, based on an analysis of what has happened with short-sighted children.

    Just listened to an interesting presentation on a related problem in development economics, where NGOs tend to overestimate the benefits of scaling up (say) a micro-credit scheme, following the successful implementation of the pilot programme. The underlying issue: Those most predisposed towards using micro-credit effectively are the very people who volunteer for the pilot. The rest of the village population may be far less entrepreneurially minded and ill-suited to making effective use of micro-credit.

  6. Yep.

    Half my job at the Urban Institute is thinking through these issues - selection, what margin we're looking at, how we have proper identification of that margin of comparison (we do some massive random assignment comparisons, but a lot of it is quasi-experimental, so we have to think through how to make that convincing). These have huge implications, but at the very least the general public should try to understand what can and can't be generalized from certain comparisons.

  7. It's gated but perhaps some people with academic library access can get to it:

    This is a paper that a colleague of mine presented recently on some of these issues - it's quite good. He refers to a quasi-experimental evaluation of DOL's High Growth Job Training Initiative (at least his presentation to us did - I assume its in here). I did most of the data work for that evaluation, and helped a lot with the conceptualization of the analysis.


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