Ben Southwood sent me an article discussing Charlie Bean’s views on NGDPLT. Let me first apologize to regular readers; much of this will be going over ground I’ve already discussed dozens of times. Unfortunately many of the leading economists now jumping into the NGDPLT debate do not keep up with the blogosphere, and hence are not aware that their arguments have been addressed numerous times.
Charlie Bean argues that the main difference between an inflation target and a nominal income growth target relates to communication – and he asks which framework would be a better way to describe policy. One argument in favour of a nominal income growth target is that there may be fewer divergences from the stated target for income growth than from an inflation target if there are a lot of costs shocks that are accommodated. But set against that, nominal GDP is more prone to revision than inflation which makes it harder to hold the MPC to account. And income growth targeting would be harder to explain to households and firms and so be less effective at anchoring inflation expectations.
NGDP targeting is actually far easier to explain to households than inflation targeting. In 2010 Bernanke indicated that the Fed was determined to raise the cost of living of the average America. People reacted with shock and outrage, and there was a firestorm of criticism. Most people associate the term “higher cost of living” with “falling living standards”—i.e. supply-side inflation. Thus they have no understanding of inflation targeting at all, which deals with stabilizing demand-side inflation. Tell the average person about the need to be symmetrical when missing the target (i.e. low inflation is just as bad as high inflation) and you’ll see a blank look on their faces. The press still reports below target inflation as “good news,” so they are no better than the man on the street. In contrast, if Bernanke has announced in 2010 that he was trying to speed up growth in the average income of Americans to boost the recovery, QE2 would have been far more popular, and the public would have had a much better idea of what the Fed was actually trying to do.
And of course the BoE trying to boost inflation when it was already well above target created all sorts of confusion in Britain, and contributed to their sub-optimal fiscal policy decisions.
I’ve dealt with the data revisions issue before. In this paragraph Bean was discussing NGDP growth rate targeting, not level targeting, hence all that needs to be said is that the central bank should target the forecast. In that case revisions are not a problem. In the case of NGDPLT, I’d refer you to this earlier post. The article continues:
Charlie Bean argues that a nominal income levels target would involve rather different policy settings to an inflation target. Under such a regime the MPC would be tasked with returning nominal income to a continuation of its pre-crisis trend line in response to economic shocks. In theory, such a target may be a useful way to influence expectations, particularly when policy is constrained by the zero lower bound on interest rates. It acts to persuade people that if there is a large negative demand shock, interest rates will be “loose for longer” than under the present inflation targeting regime because they know the MPC will need to close the shortfall in the level of nominal income. Those lower expected future interest rates directly boost demand today.
This is true, but slides over the more important point—NGDPLT greatly moderates the initial move away from the trend line.
But the policy also generates higher inflation in the future, thus producing a second source of downward pressure on future real interest rates, which also raises demand today. Charlie Bean shows that in standard economic models this expectations channel can be powerful enough to deliver a substantial economic benefit when interest rates are at their zero floor. But he raises three real world caveats.
First he notes that in standard model simulations, a large negative demand shock generates deflation, which drives up real interest rates, further depressing demand. But in contrast, inflation in the UK has averaged well above target in recent years, suggesting there have been negative supply shocks as well as negative demand shocks. The advantage of a levels target for nominal income is less clear under those circumstances.
Not at all, if NGDP growth was well below target in Britain then a level targeting regime would have led to a smaller drop in AD, and hence a milder recession.
Second, a nominal income level target will mean tolerating periods of higher than normal inflation. Models don’t capture the risk of such a policy. Demand may be depressed as savers worry about the real value of their assets. Long term inflation expectations may get pushed up and it could be costly to bring them back under control.
As long as NGDP growth is on target, higher inflation expectations will not feed into higher nominal interest rates or higher nominal wage rates. Hence the higher inflation will not distort the capital and labor markets in the way that standard models suggest. The problem here is that what is usually regarded as the welfare loss of inflation, is on closer inspection the welfare loss of excessively high and unstable NGDP growth.
Third, holding interest rates at very low levels for a long period, which would be required by a nominal income levels target, may generate credit / asset price booms and financial imbalances and so threaten financial stability. Overall one needs to be cautious about committing to loose monetary policy long after the economy has normalised.
Low nominal interest rates are a sign money has been tight. That’s why the lowest interest rates in the world are in Japan, which has the lowest NGDP growth. The developed country with the fastest NGDP growth is Australia, and it has the highest interest rates among the major developed economies. So Bean has things exactly backwards. In any case the central bank should not worry about bubbles. As we saw in 1987, bubbles do not cause significant problems as long as the central bank keeps NGDP growing at a steady rate. Problems with the financial system should be addressed through regulation, not driving all sectors of the economy into a recession in order to punish the one sector that is misbehaving. The Fed tried that in 1929—to say it didn’t work out too well would be an understatement.
Charlie Bean concludes that it is sensible to review the monetary framework from time to time, but there is a danger in expecting too much from monetary policy.
I agree. I think Britain has significant supply-side issues that need to be addressed.
“The Great Recession of 2008-9 was unlike earlier policy-induced downturns….it should not be surprising that the recovery since the middle of 2009 has been so muted”. Central bank policies “cannot – and should not seek to – prevent the necessary de-leveraging and rebalancing of production away from non-tradables towards tradables. That is a real process that takes time and means that the recovery is likely to remain somewhat subdued by historical standards.”
Yes, real adjustments need to occur, but they can occur most effectively against the backdrop of stable NGDP growth. The recovery process that Bean wishes to see is exactly what NGDP proponents favor, and yet is nearly the opposite of the policies that have actually been adopted. In the US the huge drop in NGDP led to a big fall in the production of tradables (and other types of-non-residential construction output), making the adjustment far more difficult and prolonged. I believe something similar occurred in Britain.