For more information, see Canada and the IMF

The Policy Challenge at Low Inflation
Presentation by Mr. Gordon Thiessen
Governor, Bank of Canada
to the IMF Interim Committee Meeting
September 26, 1999

Thank you Mr. Chairman. I am pleased to introduce this topic, hopefully in a manner that stimulates a free-ranging discussion.

The chapter in the IMF's World Economic Outlook (WEO) focuses on the costs or problems of operating at low inflation. As a long-time central banker, Mr. Chairman, you will not be surprised that I would like to begin with a brief reminder of the large benefits that low inflation brings to us. The WEO chapter wonders why low inflation has not eliminated financial fluctuations and difficulties. However, as I see it, low inflation has helped us cope with a serious financial shock. This has certainly been evident in Canada recently-- low inflation helped us to get through the impact of the Asian crisis on commodity prices.

More generally, there are the economic benefits in terms of overall better economic performance. These benefits are well-known and broadly accepted these days. And they are also becoming more and more evident, with perhaps the recent U.S. experience the most interesting. While the situation in the US is not without risks, the conjuncture there of low inflation, along with the strongest growth and the lowest rates of unemployment in a very long time, has impressed us all.

But there are also significant benefits from low inflation for the process of making monetary policy decisions -- particularly when expectations of low inflation are well anchored, by which I mean that the public firmly expects low inflation to persist.

If inflation expectations are well anchored and therefore do not change quickly in response to shocks, it is much easier for monetary policy to deal with such shocks. For example, when facing a shock where demand is suddenly much stronger than expected, you have more time to judge how large and how persistent that shock is likely to be. By contrast, in an inflationary environment, monetary policy has to react much more quickly and strongly in order to avoid a wage/price spiral taking hold. The same holds true with a price shock --say, a major rise in energy prices. Obviously, you have no choice but to absorb the initial effects on your price level. But in a secure low inflation environment, monetary policy canbe less concerned about second round effects on prices and wages more generally.

There is also the possibility in a low inflation environment for monetary policy to tentatively accommodate higher levels of demand when there is uncertainty about the level of potential at which the economy can operate. If the central bank miscalculates, and demand is too strong relative to supply capacity, policy can pull back without the same disastrous consequences as would occur when inflation was already high.

These are some of the benefits, but let me now turn to some of the key policy challenges at low inflation.

While low inflation gives a central bank more room to manoeuvre, one must be very careful not to become so relaxed that serious mistakes are made. For one, when inflation is low and price expectations are well anchored, there is a risk of thinking you can run the economy with excess demand on some sort of continuing basis. As well, conducting monetary policy in a low inflation environment requires a concern about excess supply leading to ongoing deflation. In other words, with low inflation, the concern of the central bank must be symmetric -- i.e., an equal concern about inflation and deflation.

This is where I believe there is a great advantage to using inflation targets with explicit upper and lower bounds. With such targets central bank must treat the top and the bottom of the range equally seriously through symmetric policy response. This symmetric behaviour will help to anchor price expectations and prevent both an inflationary and deflationary spiral.

Besides this need for policy makers to avoid becoming complacent, there are other important challenges facing monetary policy in a low inflation environment. The IMF raises three in the WEO.

One is whether a downward rigidity of nominal wages will increase unemployment at low inflation. This is an important issue requiring further research, but I must say that the evidence from Canada leads me to believe that it is less of a problem than suggested in the IMF document. Moreover, looking forward, I expect that behaviour will evolve in an environment of low inflation in ways that will promote greater wage flexibility. With low inflation, the future is clearer to workers, and they will be less concerned about being left behind because inflation is undermining the value of their wages. Some evidence of increased wage flexibility can be seen in the trend toward greater use of variable forms of compensation, such as bonuses, profit sharing and stock options in some countries.

Another challenge raised in the WEO, and clearly evident in the situation facing Japan, is how monetary easing can occur to offset a negative demand shock when nominal interest rates are at, or not that far from, zero. Let me make two points to stimulate discussion. First, a desired monetary easing could be achieved through other channels. For those of us with open economies and a floating exchange rate, the exchange rate is a major channel for monetary policy. Second, other policies can be called upon. Recent improvements in fiscal positions in many countries give policy makers greater flexibility to use counter-cyclical fiscal policy. One of the costs of high deficit and debt levels in the past has been an inability to allow automatic stabilizers to work. Moreover, we need to remember that it was in the period of high inflation that some of the largest economic cycles in the post-war period occurred. With low inflation and more damped cycles, there is less likely to be a need for the stimulus of negative real interest rates.

A final issue raised in the WEO is the challenge for policy of asset price movements and in particular whether monetary policy should respond to changes in these prices. Clearly, both sharp and sustained movements in asset prices can pose difficult problems for the conduct of monetary policy. We have seen from recent Japanese experience that when an asset price bubble occurs and then bursts, it can have severe economic consequences. On the other hand, the 1987 equity price correction had less effect on economies than anticipated, and many of use made an error in responding to it. However, while it may be too early at this point to conclude, I do not see a clear link between asset price volatility and low inflation. In other words, I do not see asset price volatility as necessarily being a consequence of low inflation.

What is important is that we closely monitor asset price movements and extract what information we can from those movements. Most importantly, we need to be on the look out for asset price movements that appear inconsistent with the underlying fundamentals of the actual and expected economic situation. But I have no neat principle or policy rule to offer as to what extent or how central banks should take assets prices into consideration.

I think I will stop here, Mr. Chairman, and give others a chance to express their views on these important issues.