Bank of Canada Inflation Targeting

It was a ho-hum announcement, and I don’t know if it generated much media coverage in Canada or not.  But last week the Bank of Canada and the Dept. of Finance announced another 5-year extension (to 2011) of the current inflation targeting regime (keep total year-over-year inflation within a point above or below 2 percent, using the core inflation rate CPIX as the more important guide for policy movements).

Here is the link for the Bank’s backgrounder on the issue, which is predictably self-congratulatory:

The Bank plans to undertake a big research program over the next couple of years, considering in particular whether it should reduce the target (perhaps to 0-2 instead of 1-3), or in fact target the price LEVEL rather than the inflation RATE.  This latter move would be bizarrely dangerous: it would require the Bank to deliberately engineer deflation (or disinflation) to offset any “errors” in which actual inflation diverged from target inflation.  (Example: suppose inflation overshot the 2% target for one year; the Bank would then need to engineer below-target inflation for one year to bring the price level back to its target.)  Given that the Bank still relies on an underlying Walrasian model in which the value of low inflation is that it allows agents to make more informed general-equilibrium decisions and actions (thus leading to a more efficient GE solution), it would be a logical (though utterly impractical) extension of their logic to adopt price level targeting.  The Bank will probably get big support for this possible change from conventional monetary theorists (David Laidler, et al.).  We on the left had better be ready to push in the other way.

Apart from preparing to counter this potential tightening of the inflation target noose, I am frankly not quite sure what heterodox economists should say about targeting.

I accept that the costs of disinflation were huge.  I reject the Bank’s view that something like 7 percent unemployment is “full capacity.”  I also reject the implict assumption that changes in unit labour costs arising from excess tightness or slackness in labour markets are the main reason that inflation either rises or falls (and I think there is potential for some interesting empirical work into the link, or lack thereof, between labour markets and inflation in the recent Canadian experience — it’s one of the weaker links in the Bank’s logical chain of causation, in my view).  I accept the argument that modest inflation can act as a valuable economic “lubricant” allowing for less painful adjustments in relative prices (including wages) — although I’ve never been convinced that you need more than 2-3% inflation to do this.  I do not believe in a stable trade-off between inflation and unemployment (though obviously for different reasons than a Friedman or Phelps would propose), so I don’t believe we accomplish anything for the workers by simply allowing inflation to rise.

But now that we’re in a low inflation environment, however, it’s probably worth trying to stay there (with appropriate flexibility and balance).  Has targeting been as successful as the Bank argues?  Perhaps it has been helpful, although for alternative, non-Walrasian reasons.  In some post-Keynesian and other heterodox models, inflation can become a “customary” variable: hanging by its own bootstraps, as it were, until some powerful force pushes it one way or the other.  Targeting may create something a bit more tangible for that “customary” variable to stick to.  This could be valuable if it helped to restrain inflationary pressures (arising from producers selling into buoyant product markets, as much as or more than workers trying to get higher wages in a tight labour market) as unemployment declines.  Mark Lavoie was highlighting some of these issues in his very interesting presentation to the PEF macroeconomic session at the CEA meetings in Montreal last spring.

On the other hand, heterodox economists in Europe (like Malcolm Sawyer and Philip Arestis) have been extremely critical of inflation targeting over there.  I wonder if it is targeting per se that has been the problem in Europe, or whether the general ossification of the European central bank and the particular circumstances of currency unification were the actual culprits.  The U.S. continues to benefit from the most pro-active and flexible monetary policy in the world; their central bank, importantly, is not constrained by a formal target (though Bernanke is warm to the targeting idea, and we all know that the Fed has an inflation target of some kind in its back pocket — right beside the NAIRU that it also stores back there).

I’ll conclude with an open question.  Is there a non-knee-jerk critique of inflation targeting that progressive economists can make in the Canadian context?  Or should, in fact, we embrace targeting as one part of an alternative monetary policy regime (one that also featured an explicit commitment to full employment and a different underlying theoretical model about what actually causes inflation)?

Insights on this point are most welcome.  I might even devote my next Globe column to this underexplored topic.


  • I share the concerns expressed by Jim Stanford. The Bank of Canada backgrounder on inflation targeting in my view shows that the hawks are gaining ground at the Bank. Previous work, well known to the Bank staff members, shows: (1) that inflation has abated, even in countries that do not have (explicit)inflation targets; (2) that deflation can have catastrophic consequences, because the central bank cannot set nominal interest rates any lower than zero, rendering interest targeting powerless; (3) that conducting open market operations that provide commercial banks with loads of excess reserves is quite useless, as illustrated by the Japanese case, thus demonstrating that monetary policy with zero or negative inflation rates can become totally powerless; (4) there are no measureable economic gains to zero inflation or price stability, relative to a situation of 2% inflation or even single-digit inflation, as underlined by Chris Ragan from McGill.

    Given all this, one is at a loss to understand what the Bank of Canada is up to. From a heterodox standpoint, it would be best for society as a whole if the Bank were to launch a research program that intended to demonstrate: (1) whether the NAIRU is really a viable concept, or just part of the textbook imaginary world, given the empirical evidence provided by meta-analysis, which shows that the best empirical studies (from a statistical viewpoint) reject the NAIRU and support unemployment hysteresis, while the worse studies, including presumably those of the Bank, tend to give support to the NAIRU (See T.D. Stanley, Journal of Economic Surveys, 2004); (2) whether it could be possible for the Bank to test the waters more than it currently does in keeping low interest rates, thus achieving lower rates of unemployment and higher rates of capacity utilization, that would provide higher standards of living for all Canadians.

  • I should like to know where Douglas Peters stands on this question, given his affiliation with Dion.


  • I strongly agree that targetting the price level is dangerous. I raised this at a meeting with Europen union economists a bit ago, and nobody had ever heard of such an idea.

    Like Jim, I can see no compelling reason to oppose the 1% to 3% target now that we have paid the price of getting there, and said as much in a CP story. That’s probably enough for relative prices to adjust without nominal wage cuts. It would seem that Pierre Fortin was wrong in arguing for a ‘natural rate of inflation’ in Canada of more than 2% such that a 2% target would require a ratcheting up of inflation (though he was certainly right about the pain of deflation.) I suspect the degree of labour market flexibility “achieved” in Canada means that even very low unemployment does not spark significant wage presssures – just witness the fact that real wages are still barely increasing. I plan to post a longer note on that tomorrow – but notes that average hourly wages outside Alberta are up just 2.5% year over year, and that union wage settlemements are in the same ballpark. A 2% target plus 1% trend productivity should allow for nominal wages to be growing by 3% a year, but that’s above the current path even at 6.3% unemployment (November.)

  • Sorry – re Fortin I meant, of course, a ratcheting up of the unemployment rate.

  • Inflation targetting is fine, just so long as it does not get in the way of the central bank achieving its other objectives — strong economic growth, low unemployment, a viable balance of payments and a reasonably stable Canadian dollar. Some of us are old enough to remember that the Bank of Canada used to target “growth rates in the money supply.” I noted to the Deputy Governor the other day that the latest Report of the Bank had two blank pages and asked “Is that where you discuss the growth rates of the money supply. Because that subject is not mentioned anywhere else?” Oh well, I guess the Bank has to have some one target.

  • The Bank of Canada by raising interest rates well above U.S. rates in the early 1990s did considerable damage to the economy. The current low inflation environment is primarily due to the China/Wall Mart effect on manufacturing and consumer prices, not inflation rate targetting.
    Central banks can readily make things worse; their ability to reduce inflation if marginal 1-3 % at most. Instead of playing god, they shoud be targeting low, stable interest rates, and forget raising rates to slow inflation.
    In reality they have abandoned their consitutional mandate as it is outlined above by Doug Peters.
    The Bank is in charge of economic policy: it sets the table for fiscal policy. If they start to raise rates it is becasue they judge the government has gone to far in stimulating demand.
    Where did the ludicrous policy of paying down debt instead of paying out employment insurance benefits come from?
    The Bank forced the government to cut in the 1995 budget, then the interest cut happened. That provoked the devaluation of the Can. dollar which got the economy growing. None of this was decided by cabinet or parliament. The cuts came from transfers, and they have had terrible consequences at the level of provincial welfare policy, and regional economies.
    Without E.I. people move to Alberta, without welfare they are on the streeet. This is not bad enought for the Bank, now they want to reduce the inflation target to 1%.
    The Bank board of directors should be representative and have some power. The top Bank officials should have to face parliamentary committees continuously to explain what they are up to.

  • Picking up on a point Duncan Cameron made, I wonder what relevance inflation targeting can have when international trade is added into the mix.

    On the import side, as Duncan notes, the broad level of consumer prices can be held in check by the huge volume of exports coming from China and other low wage countries.

    On the export side, we often hear about capacity utilization in the Canadian context. But if we are exporting a large share of that production, why should we care if tight capacity has the potential to raise, or is in fact raising prices?

    And even if it is, does this not spur new investment to increase capacity, something we should find desirable? Empirically, investment is driven more by growing demand than supply-side measures (tax cuts, deregulation), so tight capacity (up to a point, anyway) should be seen as a good thing if it means more investment. The Bank may well be undermining our productivity performance as a result by being overly hawkish on inflation.

    Apart from the broad magnitudes of exports and imports as a share of GDP, I’ve not seen an analysis that considers these effects, and what they mean for inflation policy. It seems to me that much of our thinking on inflation comes from a closed economy model, though I am no expert in this area.

  • I just read a very useful and accessible article by Malcolm Sawyer and Philip Arestis in the 2006 Cambridge Journal, critiquing inflation targeting and the “new consensus in macroeconomics” (NCM). They highlight the assumptions underlying the NCM approach:

    1. supply shocks are temporary and net out; the real economy has an autonomous tendency to settle at “full employment” (ie. a real equilibrium with stable inflation)
    2. inflation is always a problem of excess demand (corollary to the old monetarist “inflation is always a monetary phenomenon”)
    3. adjusting the interest rate can always sufficiently adjust aggregate demand to eliminate any excess (and hence reach a stable inflation point)
    4. the interest rate at which excess demand = 0 is akin to Wicksell’s “natural rate,” and this real (equilibrium) interest rate is knowable, stable, positive, and not inconsistent with an open economy’s exchange rate constraints (this latter point is highly relevant in Canada).

    They critique the NCM targeting approach on several grounds:

    1. interest rates affect inflation weakly, slowly, and unpredictably (their rule of thumb: it takes 100 basis points of interest to move inflation 20-30 basis points, and only after 18-24 months)
    2. changes in expectations, the credibility of monetary policy, and the composition of expenditure (and hence its interest-sensitivity) will alter the effectiveness of interest rate adjustments
    3. sustained cost-push and other non-demand inflation is possible, and demand-management won’t necessarily be effective in dealing with it
    4. other tools (they mention particularly counter-cyclical fiscal policy) may be more effective in controlling inflation
    5. the real equilibrium interest rate (if it even exists) will vary over time with technology, savings propensities, etc.
    6. interest rate adjustments affect real variables, even on a sustained basis; monetary policy is not neutral.

    I agree with most of this argument, with a couple of caveats. In my experience interest rate fluctuations are indeed a powerful (if blunt) weapon for controlling aggregate demand. And given the current highly politicized nature of fiscal policy, I doubt it could play as timely & effective counter-cyclical role as Sawyer & Arestis imagine (though institutional changes, such as restoring the power of automatic stabilizers, would help).

    I ended up writing my Dec. 18 Globe column on the Canadian dollar, which has finally, thankfully (and predictably) started coming back down under weight of the wicked goods-sector recession which its appreciation has sparked. I didn’t address the targeting debate explicitly, but did point out:

    * deferring oil sands developments would have been more effective in restraining what limited inflationary pressures we have faced in recent years than the rate hikes
    * the fact the Bank was driven solely by the inflation target in its interest rate decisions meant it missed a chance to prevent this goods-sector recession, which will have lasting and significant consequences for our productivity and industrial structure

    For this latter reason, I think I am coming down against targeting per se. It is not the 1-3 percent level of the inflation target that is the problem here, but rather the regime of targeting which leads (or even compels) the Bank to ignore everything else (except insofar as those other factors affect its projections of inflation) in its actions. Doug Peters’ point above that targeting is OK as long as the Bank keeps other goals in mind, will not convince NCM advocates who will argue that the target is only meaningful if it is absolute and supercedes other goals.

    Regarding other points made above:

    * I am sceptical that changing the composition of the Bank’s board of directors would meaningfully alter its behaviour, without an explicit change in the Bank’s mandate and marching orders.
    * I still think there may be some interesting heterodox story about the value of anchoring expectations in a targeting regime. I always ignored monetarist arguments about the importance of inflation expectations, on grounds that economic agents have to have real economic POWER for their expectations to matter to anything. Yet perhaps there is a way in which a pseudo-target could help to anchor expectations in a useful manner, which would allow the economy to get closer to full employment. But this would have to be credible, to be powerful, which takes us back to targeting. (Like mainstream economists, you can see that I, too, have at least two hands!)

    In sum, I think we are still a long ways from having a comprehensive, convincing left story about the problems with the current NCM regime, and what an alternative would look like. In the meantime, however, we face the more immediate task of trying to short-circuit the hawkish turn at the Bank of Canada (and its continuing flirtation with price-level targeting).

    Further comments are most welcome, and thanks to everyone for their input which I have found interesting and valuable.

Leave a Reply

Your email address will not be published. Required fields are marked *