Publications and Research

1997 conference

Price Stability, Inflation Targets and Monetary Policy

Proceedings of a conference - May 1997

Table of Contents

Session 5 - General Discussion

On the Costs and Benefits of Price Stability (PDF, 150 kb)
by Richard Black, Donald Coletti, and Sophie Monnier

Discussion: Steven James (PDF, 13 kb)

Prepared by Sheryl King and Geneviève Verdier

In their analysis, Black noted, he and his co-authors had intended to consider an Akerlof, Dickens, and Perry-type nominal wage floor. As they began to formalize this intention, however, they became convinced that modelling the effects of a nominal wage floor requires a stochastic environment with microeconomic structure, while QPM is a deterministic macroeconomic model.

In response to James's comment concerning the welfare effects of an increase in the level of government debt, Black indicated that they had not controlled for the speed at which inflation fell to the new target. He argued that inflation did not reach its new steady-state level quite as quickly because of the extra fiscal stimulus when the debt-to-GDP ratio was allowed to rise. He acknowledged that the welfare implications might have been reversed if they had controlled for the differences in the speeds of disinflation. Black also responded to the comment that the conclusions of their paper differed from those of Macklem, Rose, and Tetlow's (1994) study using QPM with respect to the welfare implications of a shock to government debt. Black agreed that this was a good point but suspected that the difference between the debt shocks in the two papers was related to model changes over the years.

With regard to the low quality of some of the papers included in the survey on the costs of inflation, Black noted his agreement with the discussants but indicated that he and his co-authors had wanted to take a neutral position in their choices and not have to choose between "good" and "bad" papers. Black also mentioned that the only papers that had been dropped were those that dealt just with countries having really high rates of inflation. Finally, he concluded, they would also have liked to use one model to measure all the costs and benefits, as suggested by Devereux, but it was difficult to have a single model with realistic dynamics and sources of the costs and benefits of low inflation.

Serge Coulombe was concerned about the implications of non-convergence as well as the consequences of a high interest rate floor in the model. Black warned him about the dangers of inferring dynamic stability results on the basis of whether the model converges or not. A technical divergence problem does not imply that the economy would diverge in reality. He agreed, however, that important consequences could result from the combination of a high interest rate floor and a low inflation target.

Cliff Halliwell agreed that the interaction of inflation with the tax system created distortionary costs but said that it was much easier to fix the tax system than to lower inflation. He commented that some Department of Finance studies showed that the benefits arising from a decrease in those distortionary costs with lower inflation did not stop at zero inflation. In dynamic general-equilibrium models, further benefits could be gained from deflation. He argued, however, that this would hardly be a good way to achieve those benefits, and suggested reducing the distortionary costs by fixing the tax system.

Charles Freedman quoted Herb Stein as saying that the public has a much better intuitive feel for the costs of inflation than economists do. He argued that we still have a long way to go in quantifying the costs of inflation, and that we have yet to see them all. In particular the debate, during the conference as well as in Bank of Canada publications such as the 1990 Annual Report, has centred on money balances, interactions with the tax system, the distortionary effects of inflation on savings-investment decisions, and the costs of protecting oneself against inflation. No one at the conference, however, had yet mentioned the amplitude of the business cycle as a cost. Freedman remarked that the recessions of the 1950s in a low-inflation environment were much smaller than the high-inflation recessions of 1981-82 and 1990-91. He argued that low inflation was beneficial because it lowered the amplitude of the business cycle, although he admitted that this was difficult to quantify. Finally he stressed that, although there were distributional effects of unemployment as discussed by Devereux, there were also important distributional effects of inflation.

Pierre Fortin expressed concern over the "crackpot" criteria used to eliminate papers in Black, Coletti, and Monnier, and indicated that he thought very little of the Feldstein estimates. He agreed with James that some attention should be paid to the Akerlof, Dickens, and Perry estimates of the costs of disinflation since this issue was so topical for the Canadian economy. Fortin also mentioned a cost-benefit analysis he had done for comments on a paper by Howitt in the recent C. D. Howe book, and suggested that the authors might find some useful insights in those comments (Fortin 1997). In particular, he noted that no attention was paid to the social costs that arise with high and persistent unemployment. Finally, he agreed with Halliwell's assertion that, to reduce the distortionary costs arising from the interaction of inflation and the tax system it would be less costly to index taxes than to reduce inflation.

Coletti agreed that it might be less costly to index the tax system but that we still needed to know the costs of operating under a non-indexed system. He also noted that, although the idea has been discussed for the last 20 years, there has been little movement towards indexation and the elimination of capital taxes. On the question of adding a nominal wage floor into their analysis, Coletti indicated his willingness to do so, but said that they would have to find a way to do it justice. With regard to Coulombe's comment on the Summers effect, Coletti argued that there were other channels omitted from their analysis that could help mitigate the problems caused by a Summers effect. In particular, even if the monetary authority could not lower nominal interest rates, it could stimulate economic activity by depreciating the currency or by monetizing expansionary fiscal policy.

Reference

Fortin, P. 1997.
"A Comment," on "Low Inflation and the Canadian Economy," by P. Howitt. In Where We Go from Here, edited by D. Laidler, 76-88. Policy Study No. 29. Toronto: C. D. Howe Institute.

Top of Page    Table of Contents