Scott Test


In the sequence I have described, the amount of reserves in the system was initially low but then rose substantially because of the asset purchases. If the level of reserves was brought way back down again, then policymakers could run their operating procedure and could raise the policy rate the same way as in the past.1

I think that approach makes a lot of sense.

However, the FOMC decided to start slowly raising the policy rate first. Policymakers were constrained by the zero lower bound when reducing the policy rate, but they were not constrained by the zero lower bound when raising it. In other words, once the policy rate was near zero, the FOMC had to use unconventional policies, such as QE, to provide further monetary accommodation when needed. But during normalization, the FOMC could adjust both the policy rate and the size of the balance sheet. The FOMC chose to use the policy rate as the primary way to adjust policy.

I still believe shrinking the balance sheet first would have been the right approach to normalization. Doing liftoff first has forced the FOMC to raise the policy rate in a world of superabundant reserves. Because reserves are not scarce as they were before the crisis, the Fed has had to adopt new operating procedures for raising interest rates.2

Effective Fed Funds Rate


  1. For more details on the operating procedure, see Williamson, Stephen. What Is Monetary Policy Normalization? Federal Reserve Bank of St. Louis Annual Report 2015. [ back to text ]
  2. To raise the policy rate, the Fed must also raise the interest rate on excess reserves (IOER) and the offering rate on overnight reverse repurchase agreements (ON-RRP). These two rates provide the upper and lower bounds of the target range for the policy rate. [ back to text ]

Additional Resources

BACK: QE3: Data-Driven, Not Date-Driven

NEXT: A Regime-Based View of the Economy