Not everyone is going to like this one:
During a recovery, unemployment seems little responsive to demand disturbances. Economic policy should focus on preventing recessions rather than trying to ameliorate their effects.
That is from the new slides/paper by Robert E. Hall and Marianna Kudlyak on the consistency of recovery from recessions, lots of evidence behind that claim, as employment recovery occurs at a remarkably consistent rate across recessions, regardless of policy response. Furthermore explanation of the micro-data mostly follows from the supply of employment, not the demand, and no that doesn’t require any kind of weird DSGE model, nor does it involve aggregate demand denialism about the initial cause of the problem. Links are here, including other papers by Kudlyak, many good papers in there, sadly these rooftops are nearly empty.