Kim McPhail asked what concrete value computable macro models can provide in monetary policy analysis and how monetary policy would have evolved in the absence of these models. Paul Jenkins illustrated the benefit of such macro models with an example using ToTEM. During the recent crisis, in a situation in which interest rates were near the zero lower bound the Governing Council of the Bank of Canada (2009) settled on the use of three alternative policy instruments: (i) conditional statements on the path of future policy interest rates, (ii) quantitative easing, and (iii) credit easing. To help the Council understand the size of the shock they were dealing with, the staff at the Bank of Canada ran ToTEM to determine the (negative) size of the overnight interest rate that would be necessary to counter a shock of the current size and return the economy to its two-percent inflation target. Hence even though it was not possible to achieve a negative policy interest rate, the Council was able to get a feeling for the duration of the period that interest rates would have to remain at the effective zero lower bound to gain the same impulse effect needed to address the shock.
Nick Rowe remarked that setting monetary policy using monetary aggregates (M) as instrument would circumvent the zero lower bound problem. He asked how the same thought experiment would have evolved if M was the policy instrument instead of interest rates. He also remarked that using a real variable, such as an interest rate, to control a nominal variable, such as inflation, cannot be done.
Jenkins recalled that during the period of money targeting by the Bank, a money supply growth rate was used to determine the movements in interest rates necessary to achieve price stability through an estimated money demand function. However, in practice, the money demand relationship broke down, prompting then-Governor Gerald Bouey to note that "M1 abandoned us, we didn't abandon M1." On the second issue, there are aggregate demand functions that clearly demonstrate that movements in real interest rates do affect aggregate demand and through this inflation. This is the core paradigm underlying the transmission process from the policy rate to the rest of the economy.
Charles Freedman asked whether the role of credit and financial variables on the economy is episodic or continuous and what sort of models would Jenkins prefer to use to deal with these issues. Jenkins recalled that for long periods of time financial variables did not appear to provide additional information to policy makers. However, Kevin Clinton's work in the 1970s and 1980s did suggest that strong growth in credit along with two or three other financial indicators were able to provide extra information.
Jenkins then called on Ali Dib to summarize the current stage of research at the Bank of Canada regarding the incorporation of credit channels into macro models. Dib reported that work is currently progressing on a theoretical model and the staff is working out the contribution of financial frictions to the cycle and their role in the propagation of shocks. The model that currently has this capacity, BoC-GEM-FIN, is not used for forecasts but rather for simulation exercises that can demonstrate the effects of different policies.