A New Measure of Monetary Policy Shocks

Available as: PDF

Central bank announcements contain both:

  • a stance on monetary policy
  • an assessment of the economic outlook

When expecting strong growth, central banks usually raise interest rates to stabilize the economy. But an announcement of interest rates higher than the market anticipated may change market participants’ beliefs, leading them to conclude the economy is stronger than they thought. When policy makers and researchers assess monetary non-neutrality, they must consider the distinction between the effect of the central bank’s monetary policy stance and the effects of a shift in fundamentals.

I measure this distinction by constructing a monetary policy shock series and applying it to the Federal Reserve’s monetary policy announcements. In particular, I examine the high-frequency movements of interest rates in a 30-minute window around the time of the Federal Reserve’s announcement. Because the announcement may reshape expectations about future monetary policy, I use changes in contract rates of interest rate futures settled in both the current and subsequent months. I then project these surprises onto the Federal Reserve’s information set, which is measured by its own economic forecasts preceding each announcement. The monetary policy stance shock is the portion of interest rate surprises which cannot be explained by the economic forecasts.

I estimate the causal effects of monetary policy on the financial market and the macro economy using the constructed monetary policy stance shock as an instrument variable. My findings are consistent with the traditional channels of monetary policy non-neutrality. A contractionary monetary policy shock will cause:

  • an upward revision in private forecasts of the unemployment rate
  • a downward revision in private forecasts of inflation
  • a decline in stock prices