A comment on Kocherlakota's suggestions for clarifying monetary policy objectives

In a recent speech (HT MT), Narayana Kocherlakota (who helps set US interest rates) makes two suggestions to clarify what US monetary policy is trying to do. The first I completely agree with. The Fed should make clear that the 2% inflation target is symmetrical. Inflation at 1% is just as much of a problem as inflation at 3%. We only need to look at the Eurozone to see the dangers of asymmetry (which in their case is explicit).

His second suggestion is that the Fed should articulate a “benchmark two-year time horizon for returning inflation to the 2 percent goal.” You can see why Kocherlakota is suggesting this change, particularly in the current context. A target which is only going to be achieved in the indefinite long term may cease to have value. However applying the two year benchmark to just the inflation target may create inappropriate pressures in different circumstances.

The UK’s Monetary Policy Committee operated until quite recently what appeared to be exactly this two year benchmark. As evidence for this, here is a chart of the Bank’s own forecast for inflation two years ahead. (The dates refer to when the forecast was made.) The inflation target was 2.5% until 2003, and 2% thereafter.

Until the Great Recession, the forecast was generally pretty close to the target. Since then, the dates on which inflation two years ahead was expected to be below target roughly correspond to dates on which the Quantitative Easing (QE) programme was expanded. (More details in this post.)

The problem with this strategy emerged in 2011. As a result of the delayed impact of the 2008 depreciation in Sterling, increases in sales taxes (VAT) and higher commodity prices, actual inflation briefly exceeded 5% in 2011. As a result, in 2011 the MPC came pretty close to following the Eurozone in increasing interest rates. (For a number of months, 3 MPC members voted to raise rates, and the remaining 6 voted for no change.) A major concern of those who voted for higher rates was that inflation would not fall back to 2% within 2 years, and that as a result the credibility of the inflation target would be damaged.

So the 2 year time horizon came close to having a very damaging impact in the UK. (Arguably it did cause some damage, because it inhibited additional QE.) Now it is of course true that the combination of cost-push shocks experienced by the UK during that period was unusual, but even if rules allow for opt-outs in exceptional circumstances, they can nevertheless exert inappropriate pressure in those circumstances. Arguably the 2013 paper issued by the UK Treasury on monetary policy was at heart a message to the Bank to no longer apply the two year ahead rule.

Luckily there is a simple way of avoiding this danger, by making a small addition to Kocherlakota's suggestion. This is to apply the two year time horizon to both the inflation target and the output gap. I can see no convincing argument why the two year horizon should not be applied to both elements of the dual mandate. The problem in the UK arose partly because the UK does not have a dual mandate. If it had had this dual mandate, and the two year horizon had applied to both elements of the mandate, then the pressure to raise interest rates in 2011 would have been much less. (Few expected the UK output gap to close by 2013, even without interest rate increases.)

There is a more general argument that is completely independent of what happened in the UK. Whatever the intention, if the two year horizon is applied to only one element of the dual mandate, there is a danger that it appears to give priority to that element over the other. So my own opinion, for what it is worth, is that Kocherlakota's suggestions are a good idea, as long as the two year time horizon benchmark is applied to both parts of the dual mandate.    

Forward Guidance is not Forward Commitment

Some recent discussions I have had with those who follow monetary policy in the UK and US, and indeed some who actually make that policy, suggest deep confusion between forward guidance and forward commitment. By forward commitment I mean a policy of committing to a future stimulus that would raise future output and inflation, in order to assist the current recovery. This is the policy that was first suggested by Paul Krugman for Japan and championed by Michael Woodford in particular. For more background, I discussed a recent paper exploring this policy here.

Why am I so confident that central banks are not undertaking this policy? First, because this policy only works through its influence on expectations. So those undertaking forward commitment have to be completely clear about what they are doing. The more opaque they are about the policy, the less effective it will be, particularly as the policy is time inconsistent. (The central bank has an incentive to change its mind once the recession is over)

Second, there is a defining characteristic of a forward commitment policy that no central bank has so far committed to, and that is to raise output above its natural rate (or equivalently to reduce unemployment below its natural rate) in the future. In short, to create a future boom. So, if that is a defining characteristic of the policy, yet no central bank has committed to do this, then as the policy has to be clear to be effective it must follow that central banks are not following the policy.

So why the confusion? First, I think there is a presumption that central bank communication will always be obscure, and that money can be made from trying to decipher their true intentions. Sometimes that may be true. In those circumstances, statements by certain policymakers that talked positively about a Woodford type policy might be relevant. However for this particular policy clarity is central. To pursue forward commitment yet to be mysterious about doing so is like announcing that you are targeting inflation but not announcing what your inflation target is.

The second source of confusion comes from focusing on inflation. A second feature of the forward commitment policy is that inflation will be above target during the boom (and perhaps before). So some have taken the part of forward guidance that says the central bank will be relaxed about inflation up to 2.5%, when their target is 2%, as indicating forward commitment. Yet that same forward guidance also features unemployment thresholds, which are above the estimated natural rate. [1] That would be a perverse thing to announce as part of forward commitment, because the whole idea is to get future unemployment below its natural rate.

A much more plausible explanation of forward guidance in the UK and US is that it is clarifying the short term trade-off the central bank will allow between inflation and unemployment. That could simply be informing the public about existing policy at a time when shocks might push inflation above target without also pushing output above the natural rate. Alternatively it could be indicating a change in that trade-off - a change in policy. In either case, the framework of that policy is entirely traditional. There is no commitment to engineer a future boom.

If I am right about this, it raises the interesting question of why no central banks during this recession have tried forward commitment. A closely related question is why no central banks have established price level or nominal GDP targets. In the case of the former this is the puzzle addressed in a recent paper by Steve Amber. To quote from the introduction: “Price-level targeting has convincing advantages, especially as a tool for avoiding the worst consequences of economic downturns. Then why haven’t central banks experimented with the regime?” He suggests that central banks are too fond of their current discretion to make this kind of commitment. If I have something interesting to say about this it will be for a future post.

[1] Here is the Fed's discussion of forward guidance. It suggests that it will be "appropriate to maintain the current target range for the federal funds rate well past the time that the unemployment rate declines below 6-1/2 percent". But at no point does it say it intends to reduce unemployment below the natural rate (between 5% and 6%) in order to raise inflation above target in the future.