Joshua R. Hendrickson
In recent years, the Federal Reserve moved to a new, flexible average inflation targeting regime. Under this regime, the Federal Reserve aims to ensure that inflation is 2 percent on average over time. This implies that the Federal Reserve will make up for periods of below average inflation with above average inflation, and periods of above average inflation with periods of below average inflation. This is in contrast to the Fed’s previous policy, in which any past deviation of inflation from its target was not factored into policy going forward. While there are potential benefits to this approach, the Fed risks creating a moving target that generates additional volatility into the economy.
Central banks have instruments, intermediate targets, and goal variables. Instruments are the things that the central bank directly controls. These are things like the monetary base (currency and bank reserves), the discount rate, and interest on reserves. The Federal Reserve can adjust the monetary base and these interest rates directly. Goal variables are economic outcomes that the central bank wants to produce. In the US, the goal variables are the inflation rate and the unemployment rate, but the goals are vague: stable prices and full employment. Intermediate targets are variables that the central bank tends to have more control over than the goal variables and often are things that the central bank observes with greater frequency than the goal variables. These intermediate targets represent important links between the direct actions taken by the central bank (changes in the instruments) and the achievement of the central bank’s goals. In other words, intermediate targets might have stable empirical relationships with the goal variables. Thus, the central bank adjusts its instruments to hit a particular intermediate target that they think will produce its intended goals. These intermediate targets tend to be things like broader measures of the money supply or the federal funds rate.
All of this sounds complicated. However, this is why having a particular target for monetary policy is important. For example, if the Federal Reserve has a 2 percent target rate of inflation, the stance of monetary policy is clear. When inflation is above 2 percent, this is an indication that policy has been too loose. When inflation is below 2 percent, this is an indication that policy has been too tight. This provides feedback to the central bank, both in terms of their job performance and the reliability of their intermediate targets in producing their desired policy objectives.
This is why the Federal Reserve’s shift to a flexible average inflation target raises important questions and concerns. One way to interpret this shift in policy is to consider it equivalent to price level targeting. Under a price level target, the central bank commits to a growth path for the price level. If the central bank wants the average inflation rate to be 2 percent, it could target a path for the price level such that the price level rises, on average, by 2 percent per year. If the value of the price index today is 100, this implies that the value of the index after two years would be 104. This could occur if inflation is 2 percent in each of those two years or if inflation is 1 percent the first year and 3 percent the next.
This is in contrast to a typical inflation-targeting regime in which any past deviation of inflation from its target is ignored in the future conduct of policy. Advocates of a price level target typically consider a “let bygones be bygones” approach to policy suboptimal. They prefer that the central bank target a path for the price level because this creates a long-run commitment device and helps to stabilize inflation expectations. Under a typical inflation-targeting regime, persistent deviations from target push the price level further and further away from its expected path without any promise to return to that path. In the short run, these deviations from the path can be costly since some economic decisions were based on an incorrect path. Ultimately, expectations will adjust to the actual path, but the intervening period is potentially costly in terms of resource misallocation.
Some consider it an open question as to whether or not the flexible average inflation targeting regime is a price level target. However, Fed chair Jerome Powell seems to have indicated that it is not, saying:
In seeking to achieve inflation that averages 2 percent over time, we are not tying ourselves to a particular mathematical formula that defines the average. Thus, our approach could be viewed as a flexible form of average inflation targeting.
Perhaps this is simply unclear language, but it does not sound consistent with a price level target. With a price level target, there would be some clear indication about where the price level was headed over the long term; the only question is how long the transition period would be. Perhaps this is what Powell meant. However, if that is what he meant, then he should say so.
If flexible average inflation targeting is not equivalent to a price level target, then it is potentially a very destabilizing policy. As I just explained, the problem with the “let bygones be bygones” approach to a typical inflation-targeting regime is that there are potential misallocations due to persistent deviations from target. Under a flexible average inflation-targeting regime in which the average is determined by discretion, the Federal Reserve is effectively aiming at a moving target.
If there is no formula for determining the average inflation rate, then how would one form expectations about the future path of the price level? This is a “let bygones be bygones” approach in which nobody knows what bygones will be forgiven. If the inflation rate is 4 percent one year, 3 percent the next year, and 2 percent the third year, what should one expect the inflation rate to be in the fourth year? The fifth year? With a price level target, there are a range of possibilities, but long-run expectations would dictate a return to the trend in the price level and therefore inflation rates below 2 percent for some period of time. Perhaps the central bank would return to the price level path. Or, perhaps the central bank would simply drop the inflation experienced in year one or year two from its calculation. Under a flexible average inflation target regime, what would happen in subsequent years is anyone’s guess.
The implications of the Federal Reserve’s decisions are important. If persistent deviations from its target can be costly in terms of resource misallocation, imagine the costs associated with simultaneous fluctuations in both inflation and the Federal Reserve’s target for inflation.
These concerns are not merely a theoretical possibility. Currently, inflation is considerably above the Federal Reserve’s 2 percent target. However, the long-run forecasts of the Federal Reserve predict a return to 2 percent inflation. There is no promise of inflation rates of lower than 2 percent at any point in the future. Adjust your expectations accordingly, if you can figure out how to do so.