Scott Sumner concludes a wry post on the EMH vs macroeconomics (e.g., “does the Fed have free will?”) with this:
1. The expected part of monetary policy has no impact on financial markets. It can still impact goods and labor markets, depending on when the policy became expected, and the duration of wage and price stickiness.
2. Because the expected part of monetary policy cannot move markets, any systematic monetary policy should not involve Fed ‘surprises’ moving asset prices. If they do, then the policy regime is non-optimal.
3. If policy is already non-optimal, and expected to remain non-optimal, then markets may be pleasantly surprised if an obscure blogger is able to make the world’s major central banks see the light and ‘target the forecast.’ That’s a good surprise, but can only occur once—during the transition from a bad to a good policy regime. After than, no more surprises. Please.”
You will benefit from reading the entire post.