Taylor Rule

The Taylor rule is a simple formula that suggests what a central bank's short-term policy interest rate should be, based on how far inflation is from its target and how far output is from its potential level. It was proposed by the Stanford economist John B. Taylor in a 1993 paper, where he showed that a rule of this kind closely tracked the Federal Reserve's actual setting of the federal funds rate from 1987 to 1992.

The rule is not a law that central banks must follow; it is a benchmark used to judge whether monetary policy is too tight or too loose. Central banks such as the Federal Reserve publish Taylor-rule prescriptions alongside their own decisions, and the rule's core lesson, raising rates by more than the rise in inflation, is built into most models used for inflation targeting.

Think of cruise control on a car. If the car is going faster than the set speed, the system eases off; if it is going slower, it adds power. The bigger the gap, the bigger the response. The driver can always override it, but the setting gives a sensible default.

The Taylor rule is a cruise control for interest rates. It starts from a "normal" rate and then adds or subtracts depending on two gaps: whether prices are rising faster or slower than the target, and whether the economy is running hotter or colder than it comfortably can. Central bankers use it as a reference point rather than an autopilot, because real economies throw up surprises no simple rule can foresee.

Taylor's original rule is:

• $i$ is the recommended nominal interest rate for the policy rate.
• $r^{*}$ is the equilibrium real interest rate, the neutral interest rate (r-star); Taylor assumed 2%.
• $\pi$ is inflation over the previous four quarters.
• $\pi^{*}$ is the inflation target; Taylor assumed 2%.
• $y - y^{*}$ is the output gap, the percentage difference between actual and potential real gross domestic product (GDP).

Worked example: with $r^{*} = 2\%$, inflation at 4%, a target of 2% and output 1% above potential, the rule gives $i = 2 + 4 + 0.5 \times (4 - 2) + 0.5 \times 1 = 2 + 4 + 1 + 0.5 = 7.5\%$. If inflation were at target and the output gap zero, the rule would give $2 + 2 = 4\%$, the neutral nominal rate in Taylor's setup.

Rearranged, the rule implies that the policy rate rises by 1.5 percentage points for every 1 point rise in inflation. Because the increase exceeds one-for-one, a rise in inflation lifts the real interest rate, cooling demand. This "Taylor principle" is widely regarded as necessary for stable inflation: if a central bank raises nominal rates by less than inflation, real rates fall, stimulus grows and inflation can run away. Studies by Richard Clarida, Jordi Galí and Mark Gertler (2000) found that the Federal Reserve violated the principle in the 1970s, when inflation became entrenched in a wage-price spiral, and followed it after 1979.

Economists have proposed many variants:

• Balanced-approach rule. Doubles the weight on the output gap to 1.0, giving more emphasis to employment.
• Inertial rules. Move the rate only part of the way toward the prescription each quarter, reflecting central banks' preference for gradual change.
• Unemployment-gap versions. Replace the output gap with the gap between unemployment and its natural level, using Okun's law; see NAIRU.
• Time-varying r-star. Use estimates such as the Laubach–Williams model, which suggested the neutral real rate fell well below 2% after 2008.
• Forward-looking rules. Use forecast inflation instead of past inflation, matching the forecast-based approach of inflation targeters.

The Federal Reserve Board's semi-annual Monetary Policy Report has included a section comparing the actual policy rate with several of these rules since 2017. Researchers at other central banks, including the ECB and Bank of England, run similar comparisons. Prescriptions differ widely depending on which inputs are chosen.

• Unobservable inputs. Both r-star and potential output must be estimated, and real-time estimates are often revised heavily, as research by Athanasios Orphanides showed for the 1970s.
• The zero lower bound. After 2008 and in 2020, many versions prescribed negative policy rates, which could not be fully delivered because of the zero lower bound (ZLB), so central banks turned to quantitative easing (QE).
• Rules versus discretion. Taylor and some members of the US Congress have argued that the Fed should be required to report against a reference rule; he contended that rates held too low in 2003–2005 contributed to the housing boom before the 2008 financial crisis. Former Fed chair Ben Bernanke and others disputed that reading and argued that mechanical rules cannot handle financial crises, supply shocks or deflation risk.
• 2021–2022. Standard Taylor-rule versions prescribed much higher rates than central banks set in 2021, which critics cited as evidence that policy fell behind the inflation surge; defenders pointed to the temporary nature of the supply shocks.

The rule is also a reference point in academic models with sticky prices, where interest-rate rules of the Taylor type replace older money-supply rules derived from monetarism.

• Federal Reserve: Policy rules and how policymakers use them
• John B. Taylor: Discretion versus Policy Rules in Practice (1993)