[mcbaby]

Skimming over Arin Dube's paper, I can't help but wonder why he doesn't address any of the lurking variables. His time frame, 1990 to 2012, represents a period of vast technological change throughout the US and the rest of the world. I admit I did not read the entire paper, but it seems Mike Konczal and a lot of us are using "causality" and "correlation" interchangeably, which doesn't seem appropriate. This is not an empirical science... lengthy extrapolations don't tend to hold their ground in real-world economics. There are so many other factors and variables working that it seems silly to make these extrapolations. Again, I did not read Dube's entire paper-—he's obviously put much more thought and time into that paper than I have this comment. With that said, that is just my .02