I think Mario's response has brought out a key problem in the way that the Bank of Canada looks at unemployment today. They certainly are aware of unemployment. They care about unemployment, and the level of unemployment influences their actions, as I think all of us have agreed. But in an inflation-targeting system it influences their actions in a very particular and peculiar way.
The bank has a judgment on what it views as the potential output of our economy. It doesn't actually have an explicit NAIRU estimate, but there is a NAIRU estimate embedded in its judgment of what the potential output is. The bank also has a theory that the dominant cause of inflation will be any time actual output exceeds what it judges to be the potential output, that is, the form of demand-driven or labour-market-driven inflation that the bank is most concerned with. They also understand that monetary policy acts four to eight quarters down the road. This is how the bank operationalizes its inflation target. It has a judgment of potential output down the road and then it tries to steer demand to a level that's not in excess of that level of potential output, to prevent the inflation that it expects would arise if you exceeded that output.
The problem is that this can easily become a self-fulfilling prophecy because labour markets, expectations, real capital capacity in our economy will adjust if everyone understands what the bank is doing. Then their assumed level of the NAIRU actually becomes the actual level of the NAIRU for the simple reason that we don't have the capacity or the ability to move beyond that.
If the bank were explicitly governed by at least a twin mandate of reasonable inflation and job growth, then it could test the waters a bit more to try to see where exactly inflation would pick up, because they would be trying to capture the benefits of the upside on the employment.