What other tribe will tell the Fed how to set interest rates, or Congress when to spend money? Mainstream macro has its discontents, but the more time you spend among the people pushing the alternatives, the more you realize how much lesser of an evil the mainstream academics represent.
We have no business throwing applied-math majors into an economics Ph.D. program. Both a liberal arts mora-philosophy B.A. or equivalent and two years out in the real world working at a job of some sort should be required.
We have no business offering a narrow economics B.A. at all. At the undergraduate social-science level, the right way of organizing a major curriculum is to offer some flavor of history and moral philosophy: enough history that students are not ignorant, enough sociology and anthropology that students are not morons, and enough politics and philosophy that students are not fools. (And, I would say, a double dose of economics to ensure that majors understand what is key about our civilization and do not get the incidence of everything wrong.)
…A first-rate undergraduate economic major will also spend due time on government failure and bureaucratic failure, and thus reach the very economic conclusion that there are substantial trade-offs, and we must pick our poison among inadequate and imperfect alternatives, even in institution design.
Pointers from Mark Thoma. Some thoughts on why there is such a focus on math.
1. Some economists really believe that the answers can be found inside equations.
2. Some economists (I think of Robert Hall) think that mathematical ability provides a reliable signal of overall intelligence, while other indicators are all more noisy.
3. It is a stable, self-perpetuating equilibrium. Once the math guys took over, they just keep giving the best jobs to other math guys.
4. Important questions in economics tend to have messy, ambiguous answers. Therefore, economists who do a bad job at answering important economic questions, or who do not even bother asking important economic questions, can do quite well for themselves.
5. Graduate students think that a class where the professor explains equations provides tangible training, while a class where a professor poses philosophical issues does not.
6. Math is a tax on bullsh*t.
I am not sure I believe that one, but I have heard it too. I had the undergrad/pre-PhD training that DeLong calls for but I still see benefits of math and statistics in economics (and psychology and politics and history). It is all about balance and choosing the right tool for the task … that varies, even for economists.
Alas, there is also cowshit, horseshit, and chickenshit. And some of the mathematically adept are stallions.
Scientism.
Personally (full disclosure: I have a math Ph.D.), I think the focus on math is for a very good reason.
If you write down your theory as an equation or three, I can do my own analysis. I can rederive the predictions you like, and also the uncomfortable predictions you don’t like. Once I derive the conclusions that don’t work, you have no recourse except to say “yep, you started with my assumptions and derived an obviously wrong conclusion”.
It’s far harder to do that with words. “That isn’t what I meant.”
Equations and math also do a great job at weeding out irrelevancies. See this comment thread on my blog, for example:
http://www.chrisstucchio.com/blog/2013/basic_income_vs_basic_job.html#comment-1124935247
The commenter “valeriekeefe” is throwing all sorts of words into the air that she thinks might bolster her case. A small amount of math reveals that at least one of those words (“worker death rate”) doesn’t change the answer at all. No point wasting words on it.
Math is a great way to force people to be clear. That’s it’s purpose, and that’s what separates economics from the rest of the social sciences.
A few comments:
1) Doesn’t seem like Cal Tech gets economics right by only teaching micro? Wouldn’t it be nice to see other universities openly acknowledge that macro is BS?
2) Even better than defining models in equations is defining them in a program. In Thomas Schelling’s model of housing segregation, he first drew out the game, and then later formalized the math. But it wasn’t until he actually attempted to program it that he realized how many assumptions needed to be made explicit.
3) And finally, just because you’ve defined something formally with math (or a program) doesn’t mean that it has any relevance or bears any resemblance to reality. DGSEs are formal. They are BS representations of reality.
7. You can become successful faster through an impressive display of mathematical economics than through an impressive display of economic history. The latter requires more time to master the literature.
It is possible to be a math-econ wunderkind, and that is the draw. Think of Sanford Grossman, Colin Camerer, Raj Chetty, Thomas Piketty (when he started out at MIT at 22), Larry Summers and of course the ur-math economist and wunderkind, Paul Samuelson.