Pierre St-Amant

This article examines whether combining forecasts of real GDP from different models can improve forecast accuracy and considers which model-combination methods provide the best performance. In line with previous literature, the authors find that combining forecasts generally improves forecast accuracy relative to various benchmarks. Unlike several previous studies, however, they find that, rather than assigning equal weights to each model, unequal weighting based on the past forecast performance of models tends to improve accuracy when forecasts across models are substantially different.

The recent global crisis was characterized by a remarkable intensity in the negative feedback process between financial sector developments and the real economy.

Model-based forecasts of important economic variables are part of the range of information considered for monetary policy decision making.

Central banks are still defining their approach to financial stability and are at an early stage in the development of useful models. The Bank of Canada's 2007 economic conference was organized to stimulate progress in the development of financial-stability frameworks. Among the highlights reported here are the discussions centred around three proposed frameworks: a contingent-claims-analysis framework, a semi-structural framework, and structural financial-stability models. Participants also reported on their experiences with stress-testing under the International Monetary Fund's Financial Sector Assessment Program and discussed the implications for financial stability of linkages among payment, clearing, and settlement systems.

At the 12th annual Bank of Canada economic conference, held in Ottawa on 4 and 5 December 2003, representatives from various public and private organizations and Bank of Canada staff discussed papers presented on three key issues affecting the financial system: financial contagion, the implications of bank diversification, and financial sector regulation. Papers on financial contagion studied the effect of globalization on Canadian foreign-asset exposures, developed a general-equilibrium model of a competitive interfirm lending market in which firms can borrow or lend, and used market-based indicators to determine the probability that contagion can be generated by interbank exposures. The papers on bank diversification focused on the links between the changing behaviour of financial institutions and risk-return trade-offs. Issues of financial sector regulation included the relationship between governance and financial sector soundness, the theoretical basis of bank regulations for capital requirements, and the implications of bank capital requirements for the transmission of monetary policy. A panel discussion provided extended discussion of the conference papers.

The third strategy employed by the Bank when dealing with uncertainty is the consideration of appropriate simple reaction functions or "rules" for the setting of the policy interest rate. Since John Taylor's presentation of his much-discussed rule, research on simple policy rules has exploded. Simple rules have several advantages. In particular, they are easy to construct and communicate and are believed by some to be robust, in the sense of generating good results in a variety of economic models. This article provides an overview of the recent research regarding the usefulness and robustness of simple monetary policy rules, particularly in models of the Canadian economy. It also describes and explains the role of simple rules in the conduct of monetary policy in Canada.