At last! After spending 57 months below the Federal Reserve’s 2% inflation objective, a key measure of price movements – the headline U.S. index of personal consumption expenditures (PCE) – rose by 2.1% year-over-year in February. In fact, this was only the fourth month PCE inflation exceeded the 2% target since it was announced by then Fed Chairman Ben Bernanke in January 2012 – and the other three times were in 2012!
So, when the inflation data was released, were champagne corks popping in the Federal Reserve Board’s Eccles Building on Constitution Avenue?
Probably not, for three reasons. First, Fed Chair Janet Yellen and her colleagues know that four months above versus 57 months below a symmetric inflation target isn’t exactly a well-balanced economic performance. Second, core PCE inflation, a better gauge of medium-term trends than headline inflation, rose to 1.8% but remains below the target (see Figure 1). And third, headline PCE inflation is likely to fall back below 2% again over the next few months unless oil prices climb at a similarly steep pace as they did from the February 2016 trough. In fact, in the eurozone, the “flash” HICP inflation estimate for March dropped to 1.5% annualized from 2.0% in February, partly due to the base effect in oil prices that will also affect the U.S. Consumer Price Index inflation data for March, due to be released on 14 April.
Many shades of inflation targeting
One year ago, I proposed that the Fed should regain lost ground on inflation by moving to price-level targeting, which would imply compensating for the past cumulative shortfall of price appreciation relative to its objective by aiming to overshoot by a similar cumulative amount in the next several years. I argued that this would help to re-anchor long-term inflation expectations, which had persistently fallen below the target and, despite some increase since the November 2016 U.S. election, have remained sub-par. However, Fed officials have repeatedly rejected price-level targeting, with the latest example provided by (dovish) Chicago Fed President Charles Evans, who said in Madrid on 27 March that the Fed should not try to make up for past deviations from target. Yet, Evans also re-emphasized that the Fed’s inflation objective is symmetric on a forward-looking basis, thus echoing the Federal Open Market Committee (FOMC) statement following its March meeting, where officials inserted the reminder that the inflation goal is “symmetric.”
Note that “symmetric” is a weaker commitment than a price-level target because it means a certain level of overshoots and undershoots are equally tolerated – that is, 2% is neither a ceiling nor a floor – and there is no promise to correct deviations once they have occurred. In contrast to a price-level target, a symmetric inflation target has no memory – bygones are bygones.
Yet, irrespective of whether a central bank has a price-level target or a symmetric inflation target without memory, it will have to take into account that when the neutral rate of interest is relatively low, the policy rate will likely spend certain periods of time – say during and after recessions – at the effective lower bound (the ELB, which is likely to be close to zero). In those times, the central bank will struggle to bring inflation back up to target because it cannot cut rates below the ELB. Thus, to achieve a 2% average inflation target over time, the central banks would have to temporarily aim at above 2% inflation in “normal” times when the policy rate is not at the ELB.
Aiming above target: how high, how long?
Once you accept this logic, the question is, how much inflation above 2% should the Fed aim for in “normal” times (when rates are not at the ELB)? The answer depends on two estimates: the expected amount of time spent at the ELB, and the expected below-target inflation rate during those ELB episodes. For example, if the Fed expected to be at the ELB for a total of five out of the next 10 years and inflation averaged 1.5% in those episodes, it would have to strive for an average 2.5% inflation rate in the other five years spent away from the ELB to meet the average inflation target of 2.0% over the entire 10 years.
Obviously, there is considerable uncertainty about how much time the fed funds rate will (have to) spend at the ELB in the future and about how low inflation will likely be during those times. In a recent paper, two Fed economists – Michael T. Kiley and John M. Roberts – estimate that the economy could spend between one-third and two-fifths of the time at the ELB if the real neutral interest rate is 1% (“Monetary policy in a low interest rate world,” Brookings Papers on Economic Activity, March 2017). Based on their simulations of inflation in ELB periods, the authors conclude that policymakers would have to strive for inflation of close to 3% when the ELB is not binding to achieve 2% inflation on average. The paper caused a bit of a stir among Fed watchers and in markets because previous papers by other Fed researchers had suggested lower estimates of both the time spent at the ELB and the required inflation overshoot when the fed funds rate is away from the ELB.
While the paper’s estimates are highly uncertain, they help to illustrate why the Fed thought it appropriate to remind markets about the symmetry of their target in March. Tolerating overshoots in “normal” times is necessary if the Fed is serious about its 2% average inflation objective and if it believes that it may be forced to bring rates back down to the lower bound in the future.
So, based on these considerations, should we safely assume that the Fed will overshoot 2% inflation as long as the fed funds rate is not at the lower bound? Not so fast.
First, given its track record and other considerations, it is not clear whether the Fed could reliably engineer a significant overshoot even if it wanted to. Wages are not very responsive to measures of slack in the economy, and inflation is even less so. In fact, national inflation may be more responsive to global rather than domestic factors due to the lengthening of global value chains over the past 20 years (see this recent Bank for International Settlements (BIS) Working Paper for details). OK, perhaps the Fed will get “help” in pushing inflation above target in the form of bumper global growth or a protectionist policy mistake by the Trump administration that leads to an abrupt shortening of global value chains, which would push intermediate goods prices higher. But neither looks particularly likely for now.
Second, at the moment the Fed rather seems to be in “opportunistic tightening” mode because financial conditions have not only remained easy but eased further despite two rate hikes within three months. The desire to get as much distance between the actual fed funds rate and the lower bound while conditions are good seems to dominate any desire to overshoot the inflation target for now, even though core PCE inflation has remained below target.
Third, but not least, given the many upcoming changes in the composition of the Federal Reserve Board, including its leadership, and of the FOMC over the next year or so, it remains to be seen whether future monetary policymakers will agree with the “symmetric” interpretation and aim at inflation overshoots in good times to build a cushion against undershoots in bad times.
Bottom line: Don’t get too excited about either the recent move of headline inflation above target or the recent Kiley/Roberts paper calculating the Fed should aim at about 3% inflation when rates are away from the zero bound. The reality is that even though inflation has been below target – by one measure for 57 out of the 61 months since the target was introduced – and even though core PCE inflation remains below target, the Fed appears to be in opportunistic tightening mode. And even if the current Fed wanted to overshoot the target, it is not clear whether it could really achieve it – nor is it clear whether a future Fed will even want to achieve it.