I got in on the tail end of the discussion of Scott Sumner’s post in which he discusses global warming. I posted a comment but it was probably too late for most people to notice.

I think the issue is more complicated than Scott seems to suggest.

Scott writes:

Theory suggests that higher levels of CO2 should raise global temperatures due to the “greenhouse effect”.


But what that doesn’t tell us is how strong the effect is. I’m not disagreeing with Scott. I’m simply saying that the effect could be strong or could be weak. If a substantial increase in CO2 led to 0.1 degree C increase in temperature, we would have little to worry about. We can’t simply look at the fact that CO2 increased and then the temperature increased and attribute the whole increase in temperature to the increase in CO2.

Take an example from the world of economics, which, of course, Scott and I are more familiar with. We posit that a substantial increase in the minimum wage will cause a substantial reduction in the number of jobs of low-skilled workers. That’s simply good economic theory. So we look at the data and see that, sure enough, a few dollar an hour increase in the minimum wage is accompanied by a substantial reduction in the number of jobs of low-skilled workers.

But that doesn’t tell us how much of the reduction in jobs is due to the increase in the minimum wage.

Similarly, it’s bad methodology to key in on one variable, CO2 concentration, and not look at other factors that could cause global warming.