- If someone suggests, as a response to a crisis, the same thing that they were implementing immediately before the crisis, we should be skeptical. I'm not saying it is impossible that staying the course is the right idea in some circumstances, but generally when situations change the appropriate response changes as well.
- If someone suggests that something will have a large effect, that effect should be evident in cursory inspections of the relevant data. I think this gets lost on scientific people a lot because we spend all of our time trying to tease out small and subtle effects, but if someone makes a claim like "demand is an important part of the economy and stimulating it should improve things dramatically" and then stimulates demand, there should be a clear change shortly thereafter. You should be able to clearly see it on the simple metrics that track the economy. Likewise, if someone says something like "high tax rates are a major obstacle to growth or business investment" then we should see a strong negative historical correlation between high tax rates and growth and business investment. If you can't see these things in the obvious data, then they aren't major factors. They may be real factors, but they aren't the big ones.
- If you claim that something matters a lot, but that it was overshadowed by some other set of effects in the historical data, then you are also claiming the overshadowing effects are as important or more important than the thing you claim matters a lot. This is a corollary of 2.
Monday, October 3, 2011
Principles of Argument and Evidence
I just have been really frustrated about the way our economic arguments have been going. Here are some principles that I feel might clarify things a bit: