Wednesday, February 28, 2007
That’s not such a trivial answer, really. If you have a subjective probability distribution concerning volatility, you should definitely revise upward the mean of that distribution. In most cases, that means acting as if there has been a moderate, but not large, increase in volatility. There is a very good chance that the actual increase in volatility will turn out either to be very small or to be quite large. Given the possibility that it could be quite large, there is reason to change ones behavior, but since that is only a possibility, the change in behavior probably should not be dramatic.
Tuesday, February 27, 2007
Is the music industry shooting itself in the foot?
In this post I’m concerned particularly with music copyrights. It is a common practice on YouTube, almost certainly illegal in many cases – perhaps in all cases – but arguably quite fruitful artistically, for non-copyright-holders to use copyrighted music in their videos. Sometimes it is an integral part of the video, and other times it is just background music. Until recently, it was also common for those who engaged in the aforementioned practice to identify, in some text format, the music in their videos. This latter practice served two functions. First, it enabled those interested in the music to find those videos. Second, it enabled those who watched the videos, but who were not previously familiar with the music, to find better copies of the music itself – often, I would imagine, for purposes of purchase. It also had the unintended function, so it turns out, of enabling copyright holders to find violations easily.
But things are changing. In the aftermath of the Viacom purge, video makers are recognizing that they do not have an incentive to identify the music in their videos. Without a text format identification of the music, each copyright holder is searching for a very specific set of needles in a large haystack. So this form of copyright violation is likely to continue – indeed to continue with greater impunity than in the past. The difference is that it will no longer benefit copyright holders by publicizing their products.
In principle, in deciding how aggressively to enforce its copyrights in this context, a holder faces a tradeoff between, on the one hand, the free advertising provided by the videos and, on the other hand, (a) the possibility that the videos substitute for the genuine musical product provided by the copyright holder and (b) the possibility that video makers would be willing to pay for copyright use if they were not able to steal it. The problem is, neither (a) nor (b) is really plausible. The audio track from a video designed to stream over moderate-bandwidth connections – in YouTube’s case, not even in stereo – is not at all a close substitute for a CD or an iTunes download. And as for (b), if you didn’t laugh when you read it, that’s probably just because I told it wrong. And then, of course, there’s the fact that enforcing these copyrights doesn’t actually stop people from violating them; it just stops people from violating them in ways that might benefit the holder.
Monday, February 26, 2007
More about Morality and Global Warming
There is almost certainly a tradeoff between economic growth and action to reduce global warming. There is room for considerable disagreement about the terms of that tradeoff, but it would be foolish to deny that a tradeoff exists. Future generations are the ones who will benefit most from economic growth. People who oppose action on global warming argue that the negative impact of such action on future generations by reducing economic growth will exceed the positive impact by reducing global warming.*
On the margin, of course, they are clearly wrong, provided that all the relevant functions are continuously differentiable. If there is some positive chance of harm from global warming, then there is some cost that is worth paying in order to reduce global warming. The optimal Pigovian tax is nonzero. It is, however, reasonably argued that some of the relevant functions are not well-behaved. If there is a “critical mass” involved in reducing global warming, the cost of attaining that critical mass may not be worth paying. Or if there is a cost to the precedent of, for example, instituting a Pigovian tax, then even a small tax may be detrimental. I don’t find these arguments convincing, mostly because I think that the expected benefits of effective action against global warming (averaging over the entire range of possibilities) are extremely high, but it is a reasonable matter for dispute (at least theoretically).
The big problem with viewing any issue as a moral one is that it more or less ends the discussion. Those who disagree with you will never be convinced, because their morality is different from yours.
*Another way to put this discussion is that economic growth, properly measured, is by definition the only thing that benefits future generations. Proper measurement would require subtracting the harmful impact of global warming. The question, then, is whether action on global warming would ultimately increase or reduce properly measured economic growth.
Not a Moral Issue
Viewing global warming as a moral issue is like viewing inflation as a moral issue. Telling people it’s immoral to overuse carbon-based fuels is like telling people it’s immoral to raise prices. WIN buttons didn’t help with the inflation problem in the 1970s, and the environmental equivalent of WIN buttons won’t help with the global warming problem. The problem in responding to global warming is one of coordinating our response, and that is precisely a political problem.
Unfortunately it’s a very difficult political problem. President Carter was ultimately able to solve the US inflation problem relatively easily by appointing Paul Volcker as Fed Chairman and letting Volcker take the heat for the necessary adjustments. By making money scarcer, Volcker’s Fed provided the incentives that induced businesses (and labor unions) to coordinate their pricing policies. Dealing with global warming will require much wider coordination (it can’t be done one nation at a time), and there is no existing mechanism to provide an independent scapegoat for the pain that may be involved.
Friday, February 23, 2007
Behavioral Economics and Economies of Scale
....behavioural economics, which the left seems to believe is a magical proof of the benevolence of government intervention, because after all, people are stupid, so they need the government to protect them from themselves. My take is a little subtler than that:
- People are often stupid.
- Bureaucrats are the same stupid people, with bad incentives.
I think Jane is wrong, partly anyway. The arguments about bad bureaucratic incentives are as strong as ever, but behavioral economics does provide a significant shift in the overall balance toward the side of government intervention. As raw material, bureaucrats are the same stupid people as the general population, but unlike the general population, they can be trained, at relatively little cost, to be less stupid.
Bureaucrats can be trained to make rational decisions in their particular areas of responsibility. Since only a fraction of the population are bureaucrats, and since the training for each bureaucrat can be limited, this training is feasible, whereas it would be prohibitively expensive to train the general population to make rational decisions in every area of their lives. In actual practice, I believe, most bureaucrats – well, many bureaucrats, anyhow – have been trained to do cost-benefit analyses and to recognize and judge the relevant costs and benefits. The average person standing in a drugstore or a grocery store has not had – and should not be required to have – the education needed to be an FDA administrator.
I see training as a special case of economies of scale that are involved in rational decision making. Even for a highly rational person, it is simply impractical – indeed, irrational – to make every decision rationally, or even to make most decisions rationally. It is costly to counter the brain’s natural irrational tendencies, and there are too many decisions to make; most of them have to be made by the not-so-intelligently designed autopilot. But when one person can make a decision for a large group, it becomes efficient to invest the resources required to produce a rational decision, even if the decision is one that a rational individual would not decide to decide rationally for herself.