Model risk is a constant danger for financial economists using interest-rate forecasts for the purposes of monetary policy analysis, portfolio allocations, or risk-management decisions. Use of multiple models does not necessarily solve the problem as it greatly increases the work required and still leaves the question "which model forecast should one use?"
In its role as fiscal agent to the government, the Bank of Canada provides analysis and advice on decisions about the government's domestic debt portfolio. Debt-management decisions depend on assumptions about future interest rates, macroeconomic outcomes, and fiscal policy, yet when a debt-strategy decision is taken, none of these factors can be known with certainty. Moreover, the government has various financing options (i.e., treasury bills, nominal bonds, and inflation-linked bonds) to meet its objectives of minimizing debt-service charges while simultaneously ensuring a prudent risk profile and well-functioning government securities markets. Bank of Canada staff have therefore developed a mathematical model to assist in the decision-making process. This article describes the key aspects of the debt manager's challenge and the principal assumptions incorporated in the debt-strategy model, illustrated with specific results.
The stochastic simulation model suggested by Bolder (2003) for the analysis of the federal government's debt-management strategy provides a wide variety of useful information. It does not, however, assist in determining an optimal debt-management strategy for the government in its current form.