JP Irving and Lars Christensen have replied to Sumner and Rowe regarding Canada. They argue that the Canadian inflation rate fell, which is a sign of a typical aggregate demand shock. They had a variety of diagrams showing the slower inflation rate.
To me, the core inflation rate has been running a bit low since 2008. Of course, Nick Rowe's initial posts didn't say that there had been no disinflation, it is just that it has been much less than he would expected given the apparent size of the output gap. Sumner's post (and diagram) did imply a much stronger statement--no change in inflation. But his point would remain that the reason why a slowdown in aggregate demand would not result in disinflation would be due to a simultaneous aggregate supply shock. The need to reallocate resources due to a change in trade can be characterized as a type of aggregate supply shock. When productivity capacity is the capacity to produce the wrong things, it is like there is less productive capacity.
Looking at Iriving's diagram and the diagram here, I am not seeing a large distinction between the GDP deflator and the core CPI. Perhaps the prices of imported consumer goods is not quite the problem I had assumed.
Leaving aside the import/export angle. If an economy simply (and suddenly) shifts demand between sectors (demand for houses grows while demand for cars falls) then during the period when the economy transitions productive capacity is likely to decline (at first the reduction in houses will not result in as many new cars being produced as when the transition is complete).
ReplyDeleteIf the money supply remains the same the price of cars would likely rise and the price of houses fall initially even if total spending remains the same and then adjust back again somewhat as the new equilibrium is reached.
For NGDPT this does not present any problem (policy can remain neutral). But for IT the policy effects will depend upon the price index used. If the index is based upon the original structure of demand then the fall in house prices will get greater weight than the change in car prices and lead to unnecessary stimulus. If a very dynamic index is used that immediately picks up changes in spending patterns then IT will also probably lead to a neutral policy.
So I agree with you that the scenario Sumner describes does not constitute a supply shock. However it does highlight that the kind of demand changes that may occur in a recession can lead to the inflation measure being skewed and leading to non-optimal policy.
My tv bill, electricity bill, house and car insurance, my groceries, my gas, my kids tuition have all increased, in fact my line of credit interest rate increased in the last year by more than 2%. I live in Canada. The only things that have not increased is my wage and the BOC CPI.
ReplyDeleteCPI is a survival index, not a cost of living index. Substitution describes survival, not maintaining a living standard. I am disappointed with the economics academia for having no one discover this.