Well the point is that it's a knowledge problem of how to best allocate resources. It's certainly possible that it wasn't an over-investment writ large just bec

Well the point is that it's a knowledge problem of how to best allocate resources. It's certainly possible that it wasn't an over-investment writ large just because a given company went out of business, but the point I was trying to make was that, because post facto you ignore sunk costs, from the perspective of someone coming around after a bust, there's all of a sudden a whole bunch of "free" infrastructure that they can build from, but this can lead you to the incorrect conclusion that it was a good idea to build all this infrastructure in the first place (and thus something you'd want to happen again). What you're actually doing is reducing the dead-weight loss from over-investment by utilizing the leftover assets, chances are it won't be a net positive, overall.