Will the Federal Reserve take a hawkish turn at its next meeting ending on 26 September? There are signs it may. Although we expect the Fed to hike rates by 25 basis points, to 2.0% to 2.25%, that’s not our concern.
Rather, there is growing evidence that some (not all) Federal Open Market Committee (FOMC) participants are de-emphasizing forward-looking projections when assessing the near-term path of monetary policy, preferring instead to focus on incoming economic data. The continued gradual pace of hikes currently prescribed by that approach is OK when the federal funds rate is still below the range of estimates for neutral policy. However, we worry that a policy strategy that continues to hike interest rates until there are clear signs of economic slowing could lead to an overtightening mistake. As we’ve discussed previously, the Fed has the tough job of balancing the risk of overtightening policy with the risk of overheating the economy.
More data, fewer models?
In mid-September, the historically cautious Federal Reserve Board governor Lael Brainard hinted at more hawkish policy in an address to the Detroit Economic Club. She said that monetary policymakers should focus on the shorter-run neutral interest rate (the rate at which output grows near potential with full employment and stable inflation) rather than the longer-run neutral rate, which also typically drives the Fed’s assessment of the stance of monetary policy. This approach, she said, argues for continued gradual rate hikes.
The shorter-run neutral rate is more sensitive to current conditions and has recently increased as growth has accelerated and the Fed has hiked rates – and it’s likely to rise above the longer-term neutral rate in the next year or two. The fundamental issue with the shorter-run neutral rate is that it’s sensitive to backward-looking economic data and is being boosted by factors that are very likely to prove temporary, including the meaningful U.S. fiscal stimulus. And while it’s possible that recent U.S. tax reforms will result in higher trend growth, we think it’s too early to tell.
Brainard’s speech echoed remarks Fed Chairman Jerome Powell gave at the Jackson Hole Symposium in late August, specifically his comments on the challenges of setting monetary policy amid uncertainty around estimated longer-run structural variables of the U.S. economy, including the longer-run neutral rate. One interpretation of the speech is that Powell prefers a monetary policy strategy that emphasizes incoming data over these uncertain variables.
Perhaps underscoring this, Powell noted that one strategy to deal with the challenge of model uncertainty was to use a first-difference monetary policy rule. Although the first-difference rule puts less emphasis on structural variables, including r* and u* – the level of unemployment where inflation is neither accelerating nor decelerating – it sets policy based on lagged changes in the unemployment rate when inflation is expected to be around target, as is the case today. When inflation is projected to be above or below target, it sets policy in part based on the number of meetings per year.
Why does this matter?
Reliance on backward-looking measures like the short-run level of r*, or monetary policy rules such as the first difference rule, risks violating Milton Friedman’s cardinal rule of central banking, namely that monetary policy acts with a lag. As former Chair Janet Yellen remarked in 2016 , half a year before starting the current hiking cycle: “The notion that monetary policy operates with long and variable lags … is one of the essential things to understand about monetary policy, and it has not fundamentally changed at all. And that is why I believe we have to be forward-looking … .”
Monetary policy lags require central bankers to set policy today with an eye toward the appropriate level of accommodation in the future. That doesn’t suggest that day-to-day data-watching should be ignored. Indeed, positive and negative economic shocks inform our views on how the economy will evolve in the future and, therefore, how to set monetary policy today.
However, views that de-emphasize the notion of monetary policy lags while emphasizing data dependence are worrisome. Backward-looking policy, which, for example, would hike interest rates when the last period’s unemployment rate declines, raises the risk of an overtightening mistake. To be sure, not all participants are de-emphasizing projections based on structural variables. For example, John C. Williams, vice chairman of the FOMC, recently discussed looking at a range of models when trying to assess unobservable variables such as the long-term neutral rate.
At the FOMC meeting that ends on 26 September, participants are likely to again revise higher their real GDP forecasts, which would suggest a stronger consensus for two more interest rate hikes this year. This would bring the fed funds rate to around 2.5% – the bottom of the range of FOMC participant estimates for the longer-run neutral level. However, after that, we believe that communication should shift toward acknowledging that monetary policy lags, coupled with the removal of accommodation we’ve seen to date, argues for a more open-minded approach as rates enter the range of estimates for long-term neutral.
Indeed, amid the absence of strong evidence of nonlinear behavior of the Phillips curve, and the existence of macroprudential tools that can offset financial sector imbalances (e.g., a countercyclical capital buffer), some continued reliance on imperfect forward-looking models appears to be a better risk-management strategy than risking overtightening with too much focus on measures that reflect the current (likely temporary) tailwinds underlying U.S. economic activity.
For more of PIMCO’s views on the complex drivers of inflation in the U.S. and globally, please visit our inflation page.
Tiffany Wilding is a PIMCO economist focusing on the U.S. and is a regular contributor to the PIMCO Blog.