Sep 12, 2018

Adopting a "Keep it Simple, Stupid" Approach to Monetary Policy

Fed Chair Jay Powell's Jackson Hole Speech Made the Case for a Simpler Path for the Fed's Future
David Beckworth Senior Research Fellow

One of the biggest challenges facing the Federal Reserve in recent years has been the apparent breakdown of the framework it uses to guide monetary policy. This framework is centered on the so-called ‘natural rate’ hypothesis, which is supposed to help the Fed avoid over or underheating the economy. Since 2015, the framework has not been working well and has created much confusion inside the Fed. Fortunately for the Fed, its problems can be solved with a KISS.

The Fed’s Failing Framework

The Fed’s framework is based on the values of the unemployment rate, the real interest rate, and real GDP that are consistent with a healthy economy operating at its full potential. These values are called the natural rate of unemployment, the natural real rate of interest, and the natural rate of output. They are colloquially known by economists as u-star, r-star, and y-star or more simply u*, r*, and y*. These fundamental values are not directly observable and have to be estimated based on some theory of the economy.

These measures underpin the key relationships that inform Fed policy. The Phillips curve, for example, is an economic model used to predict the inflation rate based on how close the unemployment rate is to u*. If the former falls below the latter, labor markets are believed to be overheating and, as a result, rising inflation should be on the horizon. On the other hand, if the unemployment rate is above u*, then disinflation should be coming.

Based on the Phillips curve, there should have been major disinflation or even deflation before 2015. There was none as core PCE deflator, the Fed’s preferred measure of inflation, has averaged about 1.5 percent annual growth since the recovery started. This Phillips curve incongruity did not bother Fed officials much and, if anything, was seen as a sign of success. The Fed had firmly anchored inflation expectations.

This confidence was shattered, however, beginning in 2015 as the unemployment rate crossed into the inflationary danger zone. The unemployment rate at this point was starting to fall below u* and Fed officials were suddenly finding their Phillips curve religion again. As a result, the Fed began talking up interest rate hikes and monetary policy normalization that year to ward off the coming outbreak of inflation. The higher inflation, though, never showed up and the unemployment rate continued to fall below u* for several more years.

The lack of rising inflation since 2015 has left many Fed officials puzzled and questioning what Minneapolis Fed President Neel Kashkari calls their faith-based belief in Phillips curves. The missing inflation concerns became the focus of the influential Economic Symposium conference in 2015 hosted by the Kansas City Federal Reserve Bank and continued to be discussed in subsequent years at the same conference. At the  Economic Symposium in 2017, for example, there was talk about the “death of the Phillips curve”. This “Great Inflation Mystery” continues to haunt the Federal Open Market Committee (FOMC), and has generated some deep soul-searching among its members.

One outcome of this soul-searching has been for FOMC members to lower their estimates of u*. Starting in 2015, the estimates for u* started falling, going from 5.60 to 4.50 percent today. This decline is consistent with, though lagging, the declining FOMC projections of r* and y*. The FOMC’s projected long-run value of r* has gone from 2.30 to 0.90 percent, while y* has gone from 2.65 to 1.90 percent.

Continually changing one’s estimates of these natural rate estimates, however, does not make monetary policy easier. As current Fed Chair Jay Powell noted in a recent speech, these estimates are not very helpful when they are constantly changing. Comparing u*, r*, and y* to the celestial stars that helped navigate ships in the past, Chair Powell said the following:

"Navigating by the stars can sound straightforward. Guiding policy by the stars in practice, however, has been quite challenging of late because our best assessments of the location of the stars have been changing significantly… the FOMC has been navigating between the shoals of overheating and premature tightening with only a hazy view of what seem to be shifting navigational guides."

The Fed Chair went on to observe that a similar shifting of the FOMC’s “navigating stars” in the 1970s created confusion and helped lead to the double-digit inflation of the 1970s and early 1980s. He is worried that these stars are just as noisy today and could lead him and his FOMC colleagues astray if they are taken too literally.

As noted by Tate Lacey, this speech echoes Jay Powell’s earlier statements that Fed officials “can’t be too attached to these unobservable variables” and should be less dependent on formal models. Moreover, the Fed Chair is not convinced that narrowly focusing on inflation as a sign of overheating is always optimal since “destabilizing excesses appeared mainly in financial markets rather [than] in inflation” in the run-up to the past two recessions. He concludes that given the uncertainty surrounding monetary policy, it is best for the Fed to take a risk management approach that moves cautiously, looks at a broad range of indicators, and is less reliant on the natural rate hypothesis.

Keep It Simple, Stupid

The Fed Chair, in short, is saying US monetary policy needs a KISS. Instead of relying on unobserved and difficult-to-pin-down natural rate estimates, the Fed should apply a “Keep It Simple, Stupid” or KISS approach to monetary policy. For Jay Powell, that means keeping an open mind to different indicators and being data dependent.

I agree with the direction the Chair wants to take the Fed, but would go one step further and contend that to really make monetary policy simple, the Fed should aim to stabilize the medium-to-long term path of a nominal variable that best meets its dual mandate. The obvious candidate for that is a nominal GDP level target (NGDPLT). This monetary regime does not require any real-time estimate of potential output or its growth rate. Contrary to some claims, this is a feature, not a bug. All it requires is that the Fed guide NGDP to it targeted growth path.

In a recent note, I show how the Fed could practically use NGDP to help assess the stance of monetary policy. I first show how the Fed could use data from the Philadelphia Federal Reserve Bank’s Survey of Professional Forecasters to create a neutral benchmark NGDP growth path. This benchmark path is essentially a weighted forecast that tries to capture the idea of “sticky forecasts”. That is, many economic decisions are made based on forecasts of nominal incomes and those decisions cannot be quickly undone. For example, a family may take out a 30-year mortgage based on an implicit forecast of their nominal income over this horizon. The actual realization of their nominal income may turn out to be very different than expected, but the individuals may not be able to quickly sell their home or move because of family commitments.

The construction of this forecast-based benchmark path does not require any natural rate estimates and is based solely on the forecasts from the Survey of Professional Forecasters. This approach indicates the stance of monetary policy was somewhat loose in the late 1990s and early-to-mid 2000s. It then turned relatively tight in 2008 and has slowly returned to neutral levels. A ‘sticky forecast’ benchmark path, specifically, indicates there is still a little bit of tightness left in monetary policy, while a ‘full employment’ benchmark suggests monetary policy is a little bit loose. Both, however, indicate monetary policy is close to neutral. 

These results tell a reasonable story and suggest this benchmark growth path of NGDP provides valuable insight into the stance of Fed policy.

A benchmark growth path for NGDP can be constructed without needing any natural rate estimate such as potential GDP.  My note goes on to show how the sticky forecast benchmark series also can be used in a Taylor rule that also requires no natural rate estimates, but is able to provide practical policy guidance when setting a target interest rate.

This analysis shows that the Fed can KISS if it so desires. All it needs to do is look beyond its current framework of natural rate estimates and instead focus on maintaining a stable growth path for NGDP. The Fed, in other words, needs to worry less about the “navigating stars” and more about maintaining the average speed of the vessel it is trying to steer. The sooner the Fed does this, the better off we will all be.

Photo credit: Jose Luis Magana/AP/Shutterstock

Support Mercatus

Your support allows us to continue bridging the gap between academic ideas and real-world policy solutions.Donate