There's been much hullabaloo over the Fed's ability to quickly remove the hundreds of billions of dollars it has pumped into the financial system after the economy recovers, but before the increased money supply sparks inflation. I and others have argued against panic, pointing to the Fed's new ability to pay interest on reserves as way to divorce money from monetary policy. And now a new study from New York Fed economists Morten Bech and Elizabeth Klee gives some quantitative support for this view:

...the results in this paper suggest that a graceful exit is indeed possible. First, our estimates suggest that, even amid high balances, the Federal Reserve can raise the effective federal funds rate when conditions warrant...[I]f the Federal Reserve raises the rate paid on excess reserves, the effective rate would move up in tandem...although the relationship will likely be less than one-to-one due to the segmented nature of the market.  Second, if the deviation between the policy rate and the effective rate is deemed to be too large, then our model suggests that the Federal Reserve can raise the effective federal funds rate even further by draining reserve using tools such as reverse repurchase agreements.

A central issue for the Fed will be how it deals with the fact that GSE's like Fannie and Freddie -- who are major players in the reserve market -- are not paid interest on their reserves. This causes the market rate for fed funds to drop below the target rate. What this means is that the Fed should be able to get the effective fed funds rate (the key short-term interest rate) back up to at least 2 or 3 percent even if it keeps the amount of excess reserves the same as it is now, but the target rate announced at Fed meetings will be higher than this level. While this isn't ideal, it does allow the Fed flexibility in removing reserves from the fed funds market.