Four fine posts from economics blogs:
Tino at Truck and Barter declares that policy on Global Warming should use two types of science: climatology and economics.
Professor James Hamilton patiently and clearly explains that causal relationships in cross-sectional data can be different from causal relationships in time series data. With some interesting comments. (Link via EconLog.)
KipEsquire--"a lawyer who doesn't practice, an investment banker who does no deals, an academic who doesn't teach, and a policy wonk who belongs to no think tank"--discusses Amtrak and tartly concludes: "So we need (taxpayer-subsidized) Amtrak because people have no options, but we also want to tweak the service (and charge more for it), because people, um, have options. Go figure." (Link via The Eclectic Econoclast.)
Finally, Professor Hamilton again, responding to the proponents of the "peak oil" hypothesis:
But here's where I get stuck. What I see many of you then in effect concluding is something along the lines of the following:
If markets are not perfect, then we should put no faith in them.
This strikes me as a very inappropriate conclusion to draw from that kind of evidence.
I'll second that!