You all know complaints I've made before about how people over-emphasize consumption behavior when talking about Keynesianism. It's an odd tendency since Keynes talked about the propensity to consume as some sort of psychological law that he didn't really attempt to derive or explain.
This post by Mark Thoma is interesting and forces people out of that box a little. He presents some leg-work by Rebecca Wilder tracking the extent of corporate savings. Usually the paradox of thrift gets talked about as a consumption-savings trade-off, but it's firms that are really sitting on the cash and it's investment that really takes the hit in recessions. The paradox of thrift itself is another area where people can have a confused approach to Keynesianism. To quote the Roberts-Papola rap, people think "savings is destruction - that's the paradox of thrift". Actually, that's not the paradox of thrift at all. Savings is not destructive at all. What's destructive is an excess of savings over investment. It's not clear why this would happen with consumer saving unless there was a large increase in the consumer saving rate. When firms save (i.e. - try to stay more liquid), though, investment is far more constrained from growing.