This has been a thought-provoking and productive day. It’s clear from the discussion that we’ve made good progress in our work, and I was pleased to hear the overall support for our approach. That said, it’s also clear that there are perspectives yet to be considered and some robustness checks of our model results that we will want to do.
Rather than summarizing all the ground we’ve covered here today, I’d like to draw out some highlights and offer my first take on where our researchers should follow up. The fact that our current inflation-targeting framework has garnered much support over the years is remarkable, but we recognize that it’s not perfect. We have a simple, flexible framework that holds us accountable, and the credibility it has brought is hard-won and invaluable.
In our morning sessions, we focused on the horse race—our side-by-side assessment of alternative frameworks. The results that Rhys Mendes reviewed are rich in information, and I commend our researchers for that. Having been part of many renewals of our monetary policy frameworks, it is evident that theory is catching up to practice.
While the models that researchers are using are still imperfect, they are richer along important dimensions. For example, we can model different behaviours in terms of inflation expectations and the fact that people differ in terms of income and wealth. In addition, we’re increasingly going beyond models to assess how well different frameworks would perform in the real world.
We saw that none of the frameworks we looked at dominate across all our criteria. However, in certain circumstances some do much better than inflation targeting. For example, targeting the price level does a better job of stabilizing prices when compared with our current framework. This is particularly true given the effective lower bound (ELB) on interest rates. That said, our experiments with real people found that these gains may not be realized because people often extrapolate from the past. Gains also come at the cost of worse outcomes when stabilizing output and distributing income and wealth.
Clearly, this means we cannot have it all. We will need to make trade-offs in our final decision on the framework with the federal government in 2021. As a policy-maker, I cannot emphasize enough how important it is to be transparent about the trade-offs that are ultimately made. Our horse-race approach, with clear criteria, will help us do just that. Already, we’re better able to clarify how our current framework works in practice. We have a single target—inflation—but we also explicitly take the output gap into account in setting interest rates. Pierre Fortin recognized this in his remarks relating to the Bank’s response to COVID‑19. The horse race also shows that our flexible inflation target is quite close to a dual mandate, in which we would target inflation and employment. It’s therefore no surprise that the differences between the two frameworks are relatively small.
The discussants found the results of our research plausible and consistent with what is in the literature.
There were nonetheless some different views on which framework we should ultimately choose. I found the discussion on how to deal with the ELB and the low inflation environment particularly interesting. One view was that we should raise the inflation target if we are in a low inflation trap. We considered this in 2016 and found that the gains weren’t big enough to make up for adverse implications for inequality and the credibility of our target. Another solution suggested was to consider “probing” or taking more risk that inflation could overshoot the target when inflation is low. Without implying any commitment to this idea, I think this approach is worth considering, particularly given the uncertainty about how low unemployment can go before it leads to inflation pressures.
Our discussants also raised several helpful technical points that are worth pursuing. For instance, both Joseph Gagnon and Pierre Fortin encouraged us to pursue alternative relationships between unemployment and inflation—through a non-linear Phillips curve, for example, and the degree of downward nominal wage rigidity. Stephanie Schmitt-Grohé suggested we should look at how the means—or averages—of key variables, such as the output gap, might change under different monetary policy frameworks. Of course, there is debate over whether monetary policy actually affects anything but the price level in the long run. Stephanie also asked us to see if our results on inequality are robust to different assumptions about fiscal policy.
In the afternoon session, we discussed policy coordination and lessons from the COVID-19 crisis. We have seen that when interest rates are at or approaching the ELB, fiscal policy—if implemented quickly—can be a powerful force in helping to stabilize the economy. Ricardo Reis and Marty Eichenbaum also emphasized the criticality of studying monetary policy frameworks in the context of the ELB and a new fiscal reality of high debt. That is why fiscal and monetary policy interaction is one of the key questions we are studying in conjunction with Finance Canada. I found Ricardo’s and Marty’s comments on the fiscal and monetary policy responses to COVID-19 interesting. They rightly raised some of the challenges policy-makers could face going forward. As a community, we need to be thinking about the implications of a new environment for our monetary policy framework.
Clearly, the recovery we are seeing is encouraging, but we know that it will take time and concerted policy efforts on many fronts to make up the ground we’ve lost. I think it’s the right time to think about how to build a strong economy in the post–COVID-19 world, particularly given challenges posed by high indebtedness. Policy coordination, while valuable, raises governance issues for central bank policy-makers, who have operational independence for good reason. The lessons from history are clear that independent central banks do a better job of controlling inflation.
Fortunately, we have learned from history. In Canada, we now have the benefit of the formal inflation-control agreement between the Bank of Canada and the federal government that is renewed every five years. The fact that it has thrived over nearly three decades and with many changes of government is testament to our joint commitment to our monetary policy framework and the economic well-being of Canadians. Marty suggested that in our 2021 agreement, we look at how the fiscal authorities could clarify their role in supporting the inflation target—potentially through enhanced automatic stabilizers, among other suggestions.
In our final session, we considered the lessons from the COVID-19 crisis and the use of alternative policy tools. Ed Devlin has come to the same conclusion we have—that the Bank’s new policy tools have helped improve market functioning and credit intermediation. We agree that it is too early to see how effective these tools will ultimately be in helping achieve our inflation target.
Now, there is some evidence from other central bank experiences in previous crises that the tools deployed at the ELB did help support the economy and raise inflation rates. But with so much going on at the same time, it is difficult to get good estimates. As Anil Kashyap pointed out, it’s harder to generate growth when key parts of financial markets are dislocated. Anil’s point about quantitative easing working through the exchange rate in small open economies like Canada’s is consistent with research that we’ve done. I agree with him that there are important identification issues in trying to estimate ultimate economic impacts.
Annette Vissing-Jørgensen walked us through some potential challenges with exiting from quantitative easing that we will want to better understand and address. Ed’s point that purchases of corporate bonds could create challenges for passive fixed-income funds adds to this.
I take Annette’s point that announcement effects can be powerful, so you will understand that I won’t reflect on the comments made today regarding monetary policy strategy.
I will say, however, that the discussion only reinforced the importance of being transparent about what we are doing and why, of commitment to our inflation target, and of continued study and refinement of the tools to make them as effective as possible.
Let me wrap up. The discussions today have clearly illustrated that we are better informed than we were when we started in 2018 but that it’s still too early for any firm conclusions. This workshop has helped refine the road map for the work that remains to be done for 2021.
Our consultations will continue, and we will continue to solicit and welcome feedback from a much broader set of people than in the past. You can look forward to more events and communications from us in the months ahead. I encourage you to keep an eye on the special page on our website dedicated to work on the renewal of Canada’s monetary policy framework.
Merci encore une fois de la part de mes collèges et moi-même pour une discussion très riche.
Thank you for sharing your views today. Even after all our work over the past two years, it is clear that there are many perspectives to be considered. It helps us immeasurably to have you hold forth with us here.