The Fed’s new policy framework remains highly controversial. But there are a number of good reasons to follow average inflation targeting.
The Fed’s adoption of average inflation targeting remains controversial. Some commentators are dubious. To them, the policy is evidence of the Fed’s manic fear of deflation. The idea of aiming for average inflation of 2% over a few years marks the end of civilisation.
They assert that the policy is highly asymmetric. While the Fed may be willing to boost inflation above 2%, they assert that the Fed would never willingly aim for inflation below 2% if it has exceeded that level for a while and the average needs to be reduced. The policy, we are told, is therefore inconsistent and implausible.
It is time for calmer commentary and a sober review of the issues. Several questions deserve attention.
First, why did the Fed adopt average inflation targeting? It did so because it was concerned that monetary policy risked losing effectiveness because of the zero lower bound on interest rates, not because of a lack of clear thinking.
Central banks set interest rates relative to a reference point. That reference point is the neutral real interest rate – the real interest rate at which output is at potential and actual and expected inflation are at target. Twenty years ago, that rate was deemed by many commentators to be 2%. If inflation was at the Fed’s 2% target, interest rates would be 4%, implying that they could be cut by 400 basis points if a recession hit. That would give a jolt to the economy.
But the neutral rate now appears to have fallen to around zero. Slower productivity growth, an ageing population and rising inequality appear to have played a role. If actual and expected inflation are at the Fed’s target, interest rates can thus be cut by only 200 basis points before reaching zero.
But if the economy and inflation are already weak when a recession hits, interest rates are likely to have been cut already, leaving little or even no scope for further cuts before reaching zero.
Second, how does averaging help? It helps because if the public expects the Fed to deliver 2% inflation on average, inflation will have to rise temporarily above 2% following a period when it has been below 2%. The public’s inflation expectations consequently rise if inflation is too low for a while, reducing expected real interest rates and stimulating the economy.
Of course, for this mechanism to work, the public must believe that the Fed will deliver on the policy, even if that means aiming for inflation below 2% from time to time. That would indeed involve a break with history, but such things happen in central banking.
For instance, if asked 15 years ago if the Swiss National Bank would ever set interest rates at -0.75% or let its balance sheet surge past one trillion Swiss francs, most commentators would have considered that inconceivable.
And few observers would have believed that the Bank of England, which achieved an average inflation rate of 8% between 1962 and 1992, would deliver average inflation of 2% in the subsequent 28 years after it adopted inflation targeting in 1992.
Thus, there is no reason to dismiss this idea out of hand.
Third, is it strange to look at the average inflation rate? No, it is not. Commentators on monetary policy often judge central banks based on the average inflation rate they achieve over some period. For instance, commenting on SNB policy we may observe that inflation averaged 2.3% in the 1990s, 1.1% between 2000 and the collapse of Lehman Brothers in September 2008, and has averaged 0.0% since then.
Furthermore, central banks are quick to say that monetary policy acts with «long and variable lags», in Milton Friedman’s words, and that they cannot control inflation in a given year. That makes it natural to manage policy on a medium term basis as average inflation targeting does.
But while average inflation targeting is less mysterious than some commentators suggest, it may not be suitable for all central banks. That is reflected in central bank practice: while it is part of the Fed’s new framework, the ECB did not include inflation averaging in its recently announced revised strategy, and the SNB is sceptical.
Why do views differ so much? Three considerations may have played a role in shaping them: whether a central bank has an objective for unemployment, how closely it wishes to control inflation, and how adverse it is to setting negative interest rates.
The Fed has a legal objective for maximum employment, a point target for inflation and is unwilling to adopt negative interest rates. These factors mean that it must avoid a sharp downswing of the economy.
The SNB, in contrast, has no explicit labour market objective, aims for inflation within a broad band, and has not hesitated to set a policy rate of -0.75% since it has deemed that necessary. It is also hesitant to review its monetary policy framework. Indeed, it started to gear monetary policy directly to inflation only in 2000, about a decade after other central banks, and has not reviewed its framework since.
The ECB is in the middle. It has no employment objective, but a point target for inflation. While it is not keen to adopt negative interest rates, it has no religious aversion to them.
The jury is still out on average inflation targeting. It may not be right thing for all central banks. But for some it will prove a useful tool for dealing with the zero lower bound on interest rates.