FindingSep 23, 2022

Perceptions about Monetary Policy

Michael D. Bauer, Carolin Pflueger, Adi Sunderam
This paper takes a new approach based on forecaster-level data to estimate a perceived monetary policy rule month-by-month from 1985 until 2021. It shows that this perceived rule varies, that even sophisticated forecasters need to learn about the monetary policy rule, and implications for monetary policy and financial markets. The Federal Reserve’s monetary policy rule is perceived differently over the policy cycle, with easings perceived as quick and surprising, and tightenings perceived as gradual and data-dependent; these perceptions affect the perceived risk of long-term bonds, or the risk premium in long-term interest rates. Learning about the monetary policy rule can therefore shed light on “conundrum” episodes, when economically relevant long-term rates appear disconnected from monetary policy.

Since it began announcing meeting decisions in 1994, the Federal Reserve has made an ever- increasing volume of information available, including detailed economic and interest rate forecasts, meeting transcripts, post-FOMC news conferences, and intermeeting speeches. The main rationale for these efforts is the idea that the public’s perceptions of monetary policy—including its goals, framework, and future course—play a crucial role in determining policy effectiveness
for the macroeconomy. Perceptions may also drive long-term rates – which matter for example for mortgage lending – by affecting the risk premium component in long-term interest rates. A substantial body of theoretical research therefore supports the notion that perception is no mere response to policy—perception also shapes policy.

However, measurement of these perceptions and how they vary over time has been challenging. While monetary policy frameworks—which include various policy tools applied at different levels and at different times—are relatively complicated, they are often described more simply via a policy rule. Researchers have typically relied on macroeconomic time-series data to analyze monetary policy rules, but these data do not capture perceptions and do not account for high-frequency changes in a policy’s parameters.

As a result, important gaps persist between what we know about the public’s perceptions of the Fed’s monetary policy rule, and how those perceptions change in response to policy actions.

To address this gap, the authors develop new estimates of the perceived monetary policy rule each month from forecaster-level Blue Chip Financial Forecasts (BCFF) data. Because these forecasters are professionals, this represents the perceived monetary policy rule of sophisticated economic agents rather than households. Please see the working paper for a full description of the authors’ methodology, but broadly speaking the authors utilize variation across forecasters and forecast horizons to estimate the relationship of Fed funds rate forecasts with inflation forecasts and output gap forecasts (the output gap is the difference between actual and potential output). This allows the authors to estimate the perceived monetary policy rule and to detect parameter shifts at substantially higher frequencies and over a longer historical period than previous work. In other words, they can more closely gauge when shifts in perception occur to infer why they occurred.

Using their new measure, the authors find the following:

  • First, the perceived weight that forecasters put on output drops toward the end of tightening cycles and monetary easing cycles but rises before, and at the beginning of, tightening cycles. The Fed is hence perceived to get ahead of the curve at the beginning of easing cycles, but to tighten in a gradual and data-dependent manner.
  • Second, forecasters appear to update their estimates of the perceived monetary policy output gap weight following monetary policy announcements in the direction predicted by rational learning, but in a gradual or even sluggish manner.
  • Third, shifts in the perceived rule explain time-varying financial market responses to macroeconomic news releases.
  • Fourth, predictable forecast errors for the federal funds rate are more likely to arise when the perceived policy output gap coefficient has increased, indicating that forecasters underestimate the Fed’s response to news, especially prior to tightening cycles.
  • Finally, the perceived output gap coefficient is negatively related to subjective bond risk premia, consistent with investors requiring lower bond excess returns when monetary policy is perceived to improve bonds’ hedging properties against macroeconomic risk.

Bottom line: The authors’ evidence suggests that changing beliefs about the monetary policy rule can explain such otherwise puzzling phenomena as when long-term bond yields decouple from changes in monetary policy rates, as occurred in 2004-2005. For central bankers, this research (and the promise of future work), offers insights into the role of perceptions and learning about the monetary policy rule, which is especially relevant for the effectiveness of monetary policy during periods when the monetary policy framework is experiencing substantial review.