Taylor Rule: Evolution, and Practice in Modern Monetary Policy

The Taylor rule and modern monetary policy

Edward Nelson’s paper presents Taylor’ rule as building a bridge between two previously incompatible monetary policy traditions during the 1990s, namely:

• Milton Friedman’s monetary policy rules tradition– emphasized benefits of policy rules but was hostile to interest-rate setting.
• Central banks’ interest-rate-setting tradition – focused on interest-rate management but viewed policy rules as rigid and impractical.

The Taylor rule represented a synthesis that brought policy rules closer to central banking practices. This formulation was significant because it excluded money stock variables entirely, representing a departure from Friedman’s monetary-growth-focused approach while embracing his concept of simple policy rules. In particular, the Taylor rule addressed key practical concerns that had historically plagued interest-rate policies:

• Friedman’s vigorous response criterion: The 1.5 inflation response coefficient embodied the “Taylor principle” – ensuring greater-than-unity responses that Friedman had argued were necessary for successful interest-rate policies.
• Output gap targeting: While Friedman remained concerned about measurement errors in output gaps, Taylor’s zero-output-gap target respected the natural rate hypothesis that both economists endorsed.

Literature revival and adoption

The Taylor rule catalyzed a major revival in monetary policy rules research during the 1990s. Key developments included:

• Empirical confirmation by subsequent studies (notably Clarida, Galí, and Gertler 2000) that showed the Federal Reserve’s responses matched Taylor’s specification
• Integration into Federal Reserve operations – Taylor rule prescriptions appeared in FOMC briefing materials starting November 1995
• Theoretical advancement- helped development of New Keynesian models and forward-looking rational expectations frameworks

From Friedman to Taylor: The Revival of Monetary Policy Rules in the 1990s
Author: Edward Nelson
From: Federal Reserve Board

Rethinking the Taylor rule

This paper by Emi Nakamura, Venance Riblier, and Jon Steinsson examines the Federal Reserve’s post-COVID monetary policy response and reassesses both the descriptive accuracy and prescriptive value of the Taylor rule in modern monetary policy. The authors investigate two fundamental questions:

1. How descriptive is the Taylor rule? – Does it accurately describe actual monetary policy across different periods and countries?
2. How prescriptive should the Taylor rule be? – When is it optimal for central banks to deviate from the rule?

Data sources, empirical analysis, and theoretical framework

• The authors employ an extensive empirical analysis spanning over 60 years of monetary policy data across multiple countries.  For the US they are using real-time Federal Reserve data from the Philadelphia Fed’s Greenbook and Tealbook records.
• For international comparisons, they utilize policy rate data from the Bank of International Settlements covering G7 countries and additional economies, along with core CPI inflation measures from the OECD.
• The empirical methodology centers on out-of-sample testing of various Taylor rule specifications, treating Taylor’s original 1987-1992 period as the baseline and examining how well the rule describes monetary policy in subsequent decades.
• The theoretical analysis employs the canonical New Keynesian model as its foundation, characterized by forward-looking Phillips curves and dynamic IS equations that capture how rational agents form expectations about future economic conditions. A key theoretical innovation is the development of a “minimalist Taylor principle” that separates the central bank’s response to fundamental economic shocks from its response to non-fundamental or speculative disturbances.

1.How Descriptive Is the Taylor Rule? 

• While the rule fits reasonably well during its original 1987-1992 sample period and continues to describe policy adequately through the Greenspan era (1987-2006), its performance deteriorates significantly in other periods. The rule exhibits particularly poor fit during Paul Volcker’s chairmanship (1979-1987).
• Since the early 2000s, deviations from the Taylor rule have become both persistent and substantial. The period from 2003-2006 saw gaps of up to 250 basis points. More dramatically, the post-COVID period witnessed the largest Taylor rule deviation on record, reaching 1,125 basis points in early 2022.
• The Taylor rule performs even more poorly when applied to other G7 countries, with only the United Kingdom showing fit comparable to the United States. For most other countries, the rule provides only sporadic periods of reasonable accuracy. Japan presents a particularly stark case, where the Bank of Japan has systematically varied policy rates much less than the Taylor rule prescribes without losing control of inflation.
• The post-COVID period reveals a striking pattern that challenges conventional Taylor rule wisdom: countries that raised interest rates more aggressively actually experienced higher inflation. “Early riser” countries increased policy rates by an average of 8.5 percentage points but saw inflation rise by 10.8 percentage points, while “late riser” countries raised rates by only 4.2 percentage points yet experienced inflation increases of just 5.9 percentage points.

2. How Prescriptive Should the Taylor Rule Be?

• When optimal monetary policy is simulated in the canonical New Keynesian model and Taylor rules are estimated on the resulting data, the coefficients consistently fall below Taylor’s original values and often below unity.
• Forward guidance justifies muted contemporaneous responses. Optimal policy relies heavily on expectations management through forward guidance, which naturally reduces the contemporaneous correlation between interest rates and inflation.
• Real-world inflationary episodes typically involve combinations of different shocks rather than the pure demand or supply disturbances assumed in theoretical models. When cost-push and natural rate shocks are perfectly negatively correlated—representing scenarios like productivity declines that simultaneously increase inflation and reduce the natural interest rate—optimal policy calls for even more muted responses.
• The ability to implement these sophisticated, shock-contingent policies depends critically on central bank credibility and anchored inflation expectations. Only institutions with substantial inflation-fighting track records can afford to deviate from simple rules without triggering expectation spirals. Central banks lacking such credibility face a less favorable trade-off and may need to adhere more closely to simple rules.
• The theoretical literature’s emphasis on the Taylor principle to avoid self-fulfilling fluctuations proves extremely fragile to modest deviations from fully rational expectations.

Beyond the Taylor Rule
Authors: Emi Nakamura, Venance Riblier, Jon Steinsson
From: UC Berkeley