Structural Breaks in Canada's Money Demand Function: Evidence from the Inflation Targeting Regime and the Covid-19 Pandemic

Abstract

Aims: This study empirically examines the validity and stability of the money demand function in Canada over the period 1961-2023, a timeframe characterized by major institutional reforms and significant economic disruptions. Study Design: Time series econometric analysis with structural break testing. Place and Duration of Study: Canadian economy, from the first quarter of 1961 to the third quarter of 2023 (251 quarterly observations). Methodology: The analysis employs quarterly data from the Federal Reserve Economic Data (FRED) database. Real money balances are measured using M1 deflated by the Consumer Price Index. Independent variables include real GDP and nominal interest rates (central bank rates). The methodology applies Ordinary Least Squares (OLS) and Generalized Least Squares (GLS) regression techniques to model the relationship. A key contribution is the incorporation of structural break analysis to account for the effects of the early implementation phase of inflation targeting (1991-1995) and the COVID-19 pandemic (2020-2023), which were confirmed to be statistically significant using the Chow test. Results: The GLS estimation, which corrects for serial correlation present in the OLS model, provides clear evidence supporting traditional money demand theory for Canada. The results indicate a statistically significant positive relationship between real income and money demand (β2 = 0.85, p-value = .00), and a negative relationship between interest rates and money demand (β3 = -0.003, p-value = .00), both consistent with theoretical predictions. The structural break analysis confirms parameter instability across the COVID-19 period, with the income elasticity dropping significantly (δ5 = -0.85, p-value = .00) and interest rate sensitivity becoming more negative (δ6 = -0.02, p-value = .00) during the pandemic. The Chow test validates the presence of a  significant structural break (F-statistic = 68.756, p-value < .001), highlighting the impact of the COVID-19 pandemic on money demand dynamics. The final GLS model explains a substantial portion of money demand variation, with all key coefficients statistically significant at conventional levels. Conclusion: The results indicate that while real income and interest rates significantly influence money demand, a structural break occurred during the COVID-19 pandemic that materially altered the money demand dynamics. This finding reinforces the critical role of precautionary behavior, transaction patterns, and economic uncertainty in shaping money demand in advanced economies. The confirmed structural break highlights the importance of accounting for major economic shocks when formulating monetary policy frameworks and using money-based indicators for policy analysis.