PGL at Angry Bear points us to this TPMCafe piece by Jared Bernstein. They both reference the Barry Schwartz piece I discussed here in my last post.
Bernstein uses the Schwartz article as a springboard for detailing the shortcomings of the economics profession, including forecasting (thus the title of this post). Two of his critiques stand out:
2. Economists are reductionists.... the world doesn’t work like the textbooks say it should.
...
3. And one reason for that is, as the NYT oped argues, we misunderstand incentives. To be specific, we exaggerate them.
I have frequently argued (e.g. in this recent post about Iranian gasoline) that the textbook model is not perfect but still useful. In my discussion of the Schwartz op-ed, I conclude that monetary incentives can have nonstandard effects in certain circumstances. We don't completely understand those circumstances and therefore more work is needed. I'm not ashamed to admit that perhaps a little more humility is also needed when considering the possibility of these non-standard effects. But it still remains that the incentive story that drives most economic models is mostly right. The question of whether we exaggerate the magnitude of these effects in our rhetoric (even when we are correct about their existence) is another matter, and I'd be more comfortable if Bernstein didn't lump them together.
I should note also that I have responded to PGL's post in the comments over there. This whole issue of responding to incentives and the usefulness of textbook economics is a worthwhile topic and generates some of the more interesting conversations on the blog. I hope to flesh out some more ideas on this over time.
I think my point - not well stated - was that income effects not only matter but may also condition substitution effects.
You're asserting a sort of Giffen paradox? If the activity is inferior enough, then the monetary incentive works backwards?
I could see this in certain cases if the offer of money offends the person. I would also accept an explanation that asserts a non-linearity. That increasing the monetary incentive from zero to anything positive changes the framing of the issue so as to alter behavior--but in those cases (as in the Swiss example Schwartz cites), it just means you have to raise the incentive.
As someone who is frequently on the other side of the argument let me explain my experience.
Economic theory is a great place to start. But it should only be the first step in your analysis.
The theory get you started in the right direction.
But next you have to look at what special factors in each situation makes this case different and will
cause standard theory to be misleading.
One of the best example is investment by the oil companies in new energy. We have a great theory that price controls discouraged capital spending on new energy in the 1970s. But you look at the data and see that there was a fantastic capital spending boom on energy during the period of price controls. Essentially the only constrain on drilling was the capacity of the oil service industry and that was expanding as fast as it could. So the theory did not work in the 1970s.
There was nothing wrong with the theory, it just that other factors massively overrode the negative impact of price controls.
Now we see energy prices soaring and should expect massive capital spending by the oil companies.
But oil company spending is very disappointing.
Is the theory wrong, or does it just give insufficient weight to the point that the top executives of the oil companies all experienced the 1980s when the oil bust nearly bankrupted their companies and they do not want to repeat that experience. So that all are very conservative on the price they use to justify need exploration .
Again, it is not that the theory is wrong. It is just that it other factors that the theory does not include also play a major role in the situation.
Or, as I tell my students: In the real world, ceteris is not always paribus.
Physics suffers from the same problem. Galileo said that two falling bodies dropped from the same height at the same time will hit the ground at the same time regardless of their mass. But if you drop a brick and a feather from your roof, the brick always hits first. Why? The answer is, of course, air resistance. Galileo wasn't wrong. But elementary physics makes some simplifications in eliminating frictions like that. And I think most people who teach physics would hold that it is useful to teach it that way before adding the frictions.
So it is with supply and demand. The model is a simplification. Sometimes the simplifications we make lead to predictions that turn out to be wrong.
At an elementary level, it may be enough to teach the simplified model and add a cautionary note that includes a simple analysis of situations like this. Just as a physics course would present Galileo's theory and an explanation that the reason it doesn't always work on Earth is because of air resistance.
At a more advanced level, the frictions can be developed and added to the model in a more comprehensive way. Just as a course in advanced aerodynamics could model the effect of air resistance with differential equations.
I have always been careful not to take the supply and demand model too literally. But anyone who claims that demand curves slope upward has the burden of proof on them.
I think the comparison to physics is interesting. I'm a sociologist by training who has studied a good deal of economics, because it overlaps with my sociological interests, and I pursue physics as what one might call a serious hobby. I think physicists, like economists, do make unrealistic assumptions in their models. In physics textbooks, there is some attention paid to the limitations of certain models and the conditions under which certain models are applicable. For example, in discussions of work, there is a model for when the force is constant and one for when the force is variable. Economists appear to do something similar in their textbooks. There is a model of price determination when the conditions of perfect competition hold, a model under monopolistic competition, one under monopoly, and several, as I understand it, when oligopoly holds. The problem, in economics seems to be when there is the move from theory to what might be called policy analysis. Take the minimum wage debate. Textbook economic theory, at least at the elementary and intermediate levels I'm most familiar with, tends to conclude that a minimum wage will cause unemployment. Yet it's obvious that not all the conditions for perfect competition are met in labor markets. The most obvious one that's not met is perfect information. Now economists have said a lot about imperfect information so I'm not saying that there is no attention to how information might not be perfect in some markets. What I am saying is that when it comes to analyzing the effect of a minimum wage and offering policy guidance based on such an analysis, most economists seem to have concluded that the conditions of perfect competition approximately hold in the real world and that this allows them to use the standard model of supply and demand. This would be like a physicist assuming that air resisitance is negligible in some situation in the real world and deciding to use the model of projectile motion without air resistance to model the motion of some object of interest. Now when I've read physicists they've been careful to spell out the conditions under which unrealistic assumptions are good approximations of the real world. I haven't seen this happen to the same extent in economics, although, as I said, I'm an "outsider" so perhaps I just haven't read enough. In other words, I haven't seen an economics textbook, and I've looked through quite a few, state relatively precise conditions under which it's "safe" to assume that perfect competition approximately holds and when it's unsafe to do so and, therefore, we should use a model more relevant to the situation. I've heard economists argue for more market forces in health care, education, and a host of other arenas on the basis of the standard model of supply and demand. But I haven't heard them make arguments, based on empirical data, as to why we should think that perfect competition approximately holds in these situations. At first glance, it would seem that the market for health care is one where the conditions of perfect competition don't apply at all so a call for more market forces in health care, based on the standard model of supply and demand, might be particularly misguided. So it seems to me that the task of economics may not be to abolish their models but to refine them by paying much more attention to the conditions under which they apply and, by doing so, their limitations.