David Stinson recently sent me an interesting article on monetary economics written by Peter Howitt. The author reminds me of people like Nick Rowe and David Laidler, as he can be sympathetic to mainstream new Keynesian ideas, but also understands the importance of older monetarist traditions. The entire paper is worth reading, but this passage on page 22-23 caught my attention:
Moreover, it is not just the policy makers that are learning from monetary theorists. Often the conduct of monetary policy is way ahead of the theory, and we academic economists often have more to learn from practitioners than they have from us. I came to realize this when I was a participant in monetary-policy debates in Canada in the early 1990s. The Bank of Canada was moving to inflation targeting at the same time as the country was phasing in a new goods and services tax. The new tax was clearly going to create a problem for the Bank by causing an upward blip in the price level. Even if the Bank could prevent this blip from turning into an inertial inflationary spiral, the immediate rise in inflation that would accompany the blip threatened to undermine the credibility of the new inflation-reduction policy.
The bank dealt with this problem by estimating the first-round effect of the new tax on the price level, under the assumption that the path of wages would not be affected, and designing a policy to limit the price blip to that estimated amount. It announced that this was its intention, and that after the blip it would stabilize inflation and bring it down from about six percent to within one percent band over the coming three years.
At the time I was very skeptical. Along with many other academic economists I thought it was foolish for the Bank to announce that it was going to control something like inflation, which it can only affect through a long and variable lag, with such a high degree of precision. To me the idea reeked of fine-tuning, and I thought the Bank was setting itself up for a fall. But I was wrong. In the end the Bank pulled it off just as planned. The price level rose by the amount predicted upon the introduction of the new tax, and then inflation quickly came down to within the target range, where it has been almost continuously ever since.
Two things struck me about this passage. The first is that economists often overestimate the problem of “long and variable lags,” especially when the goal is to stabilize a nominal aggregate. If the policy is credible, lags do not prevent the central bank from hitting short term targets, as the short run is strongly influenced by expected longer term outcomes (as Woodford has shown.)
The second thing that struck me is that this incident seems to provide some support for a policy of wage targeting, rather than price level targeting. In earlier posts I discussed Earl Thompson’s legendary (well it should be legendary) 1982 paper where he proposed a monetary regime with the dollar convertible into contracts linked to the future aggregate wage rate. George Selgin has recently put his long essay on productivity norms onto the internet, and I plan to do a post on “Less Than Zero” in the next few weeks. For now I’ll simply point out that one of the two policies Selgin considers, a labor productivity norm, would effectively stabilize nominal wage rates, but allow some fluctuations in the price of goods and services.
Obviously the Canadian example was an expedient, not the sort of permanent policy regime envisioned by Thompson and Selgin. But if it really did work as well as as Howitt indicates, why not make it permanent?
I have always thought a wage targeting regime seemed optimal in principle, but ended up advocating NGDP targeting for a variety of reasons. I was worried that nominal wage targeting would not be politically acceptable, and also worried about measurement issues. Some jobs are compensated on an hourly basis, some on a yearly basis.
BTW, Howitt cites a 2004 book by Laidler and Robson for his information on Canadian monetary policy.
PS. This is the 7th post that I did today. I’ll take a break and come back later with posts on Less Than Zero, and also a promising new way to teach AS/AD.
PPS. Several times today I have tried and failed to link to PDF files. Does anyone know what I am doing wrong?