Should the Fed Raise Rates Now to Combat Future Recessions?

Many macreconomists believe that the Fed needs to keep raising rates now so it can lower them in future to combat a recession.  See, for example, my colleague Martin Feldstein’s recent op-ed in the WSJ.

I have always been puzzled by this argument; it seems to assume an asymmetry in the effects of raising versus lowering rates. 

Specifically, if raising rates now will hurt the economy just as much as lowering them later will help, why is the net effect beneficial? To a first approximation, one might expect a symmetric effect, which makes the standard case for higher rates now unconvincing.