Alexander Schibuola directed me to a recent speech by Fed Vice Chair Stanley Fischer:
Committees and rules may appear to be in opposition as approaches to policymaking. One might even argue that if a central bank ever converged on a single monetary rule, there would be no need for a monetary policy committee. In practice, the Fed operates through a committee structure and considers the recommendations of a variety of monetary rules as we make monetary policy decisions. Our decision is typically whether to raise or reduce the federal funds rate or to leave it unchanged. Committees can aggregate large amounts of diverse information--not just data, but also anecdotes and impressions that would be hard to quantify numerically. Good committees also offer a variety of perspectives and underlying economic models for interpreting the economy. In contrast, a policy rule, strictly defined, is numerical and constrained to a single perspective on the economy.
Alex noticed that this is a good argument for using markets to set monetary policy.
You can think of markets as a sort of super committee with 7.3 billion potential members, instead of 12. Markets are privy to even more useful anecdotes and impressions than is a committee of 12, and markets have even more perspectives and underlying models.
So why not have the Fed set a 4% NGDP target, level targeting, and offer to sell unlimited NGDP futures at 5% and buy unlimited NGDP futures at 3%? In that case, the rule would be that the Fed is forced to put its money where its mouth is, anytime their policy views sharply diverged from the market consensus on expected NGDP growth. As a practical matter, the huge mistake of late 2008 would have been impossible under my "guardrails" approach because investors like me would have gone short NGDP futures at the 3% price, as it was very clear we were going to have sub-3% NGDP growth in 2009. (NGDP actually fell by roughly 3% from mid-2008 to mid-2009.) If the Fed didn't respond adequately, their losses would have been massive.
In this vision of monetary policy we have the best of both worlds. The "wisdom of crowds" that you get from decision-making by committee, as well as the rigorous constraints on monetary policy imposed by rules.