Here.
For some time now, I've been watching the blogging fisticuffs between Kathy G. and Megan McArdle with interest. Partially because I enjoy watching smart people beat the crap out of each other, and partially because I think their fight has larger implications than either of them realize.
G. has written a number of excellent posts on the need to stop looking at labor markets through the lens of the perfect competition model; rather, G. argues, we ought to use a monopsony model. McArdle, on the other hand, defends the orthodox competition model.
I should say now, I'm undoubtedly bastardizing both view points. I don't think that McArdle is arguing that we use the over-simplified ECON 101 model, and I'm probably painting G.'s views as more radical than they actually are. But I'm trying to set up a rough dialectic, not actually give the nuances of each side's position. I'm not doing this out of academic dishonesty, rather, I'm doing this because even if I did give the nuances, my point would still hold. And, since I'm trying to give a meta discourse argument, it's easier to see it through this black-and-white lens.
Also, I'm a poor student of economics.
Here's what G. wrote:
I have nothing against models in general. I do, however, have a particular issue with the perfect competition model of the labor market. I think monopsony mostly (but not always) is a more useful concept for understanding labor markets. And it happens to be every bit as much of a model as the perfect competition model is. It's a better one, though. A labor market which includes frictions (monopsony) is not only far more plausible, but the predictions you get from that model are a better match for the empirical facts on the ground.
There are a couple of points I'd like to highlight from this passage. First, is the rather obvious (but important) point that some models are better than others, but no model is perfect. If you asked any economist worth her salt, she would tell you that. The problem is, as often as not, even if that economist knows that, she's still going to put it forward not as a model or things, but as an approximate representational picture of the world around us.
And even if the economist herself doesn't do this, the public sphere will interpret it as such.
The second point is that G. correctly wrote "are a better match for the empirical facts ...." I happen to agree with G. in this case that Monopsony is a better model, but I reject her use--and I might be straw manning her argument here--of this as the only or at least most important unit of measurement.
Models in economics, like their scientific counterparts, are also chosen based on their network unification (degree to which they adhere with other models and theories), their fecundity (ability to produce new questions), their progressiveness (ability to produce novel facts), and a host of other--mostly explanatory--concerns (simplicity and so on).
It's too easy to ask that a model be a relatively good fit (or even a better fit) than its competitors. As has been pointed out by every philosopher of science since Popper (and even by Popper himself) a few ad hoc adjustments here and their, and you've got a model that can describe anything.
But, as I'm not that familiar with G. and McArdle's argument, I'm going to assume that when G. talks about "better fit," she doesn't just mean it in some kind of Popperian way, she means it in terms of explanatory goodness as well.
But, even with this caveat, G.'s argument highlights the way philosophy of science has been ghettoized to the degree that aspiring sciences--like economics--are unaware of or don't care about the epistemic norms that make for good theory choice.
Look, I'm not a freshman philosophy major who just discovered the continental school; I don't think that underprivileged groups are epistimically better well-off; and I'm not a raging Kuhnian. I do, however, think that there are more epistemic (cognitive) values than just accuracy, and I do think that one has to be aware of the non-cognitive values that will, invariably, work their way into both theory formation and theory choice.
The problem is, economics (and the rest of the social "sciences"), in their zeal to be taken seriously, have ignored, for the large part, the epistemological questions that have plagued knowledge-gathering activities. The real problem with McArdle isn't that she holds to an empirically inadequate model (though, she does) and it isn't even that her arguments can be lame ("trivial," Megan, really?); the problem is she, like most economists have insulated themselves from other discourses and refuse to see that their models are highly anti-realist.
Take, for example, McArdle's support for school choice. I'm willing to grant (for the sake of argument) that it would make sense from a purely economic standpoint to support school choice. But I refuse to accept the background values that go into that economic view.
I don't think that, for the most part, people are fungible. And I don't think it's a good idea for government to treat people like clients. I'd hold, in fact, that the erosion of our social capital and institutions is a result (in part) of this kind of client-based thinking.
To quote JFK "ask not what your country can do for you [client], but what you can do for your country [citizen]." As cliche as that quote is, the fight over welfare "reform," vouchers, and a whole host of other economic issues has moved away from the important point about our social contract that JFK was making.
As any of my more economics-inclined friends will tell you, there is no love lost between me and the field. And to be honest, part of that may be a result of a certain godawful Professor. But, I do believe that economics has much to tell us about the outside world--provided you take into account the "big picture." i.e. relevent congnitive and non-cognitive values for theory choice.
|