Monetary Policy Analysis and its Contemporary Challenges

This thesis contains three essays on the empirical analysis of monetary policy. While the subjects are diverse, they all share the goal of providing for a thorough, data-driven analysis of critical policy developments related to communications from North American central banks. The first chapter examines the effectiveness of central bank communications as a policy tool. To evaluate this otherwise qualitatively-oriented policy channel, a new dictionary of central banking sentiment is developed using natural language processing. This dictionary aims to capture the relative prevalence of positive (contractionary) versus negative (expansionary) words used in discussions of the monetary policy landscape. It is then applied to a large sample of news articles, where sentiment scores are computed and adopted in two forms of empirical analysis. The first form of analysis utilizes these sentiment scores in a high-frequency event study, which indicate that positive communication surprises lead to increased interest rates across various horizons on the yield curve, along with an appreciation of the Canadian dollar relative to other major currencies. The sentiment measure is also employed in a lower-frequency analysis, where the average score across all articles is computed on a monthly basis. VAR estimates support the findings from the high-frequency event analysis and allow exploration of other outcomes available only at a monthly frequency. The analysis suggests limited direct evidence of links between communication shocks, prices, and real measures of economic activity, except for the real estate market. In the second chapter, we profile an essential case study that emerged during COVID-related monetary stimulus, where central banks sought to dismiss concerns about rising inflation as "transitory." This chapter focuses on the United States and develops a separate tailored dictionary that is used to quantify the degree of belief (or disbelief) in the transitory inflation signal. It analyzes news articles and tracks changes in sentiment-derived signal credibility over time, revealing that overall levels of credibility declined as positive inflation surprises persisted throughout 2021. This measure is then adopted within the framework of a daily VAR model, showing that the signal credibility measure declines significantly to positive inflation surprises and that market-based inflation expectations rise even at extended horizons in response to negative shocks from the credibility measure. The final chapter explores the potential intersection between economic inequality and monetary policy in Canada. In the first exercise, a macro panel exercise reveals a "U-shaped" effect on income sourced from labour, meaning that expansionary policy benefits the bottom and upper ends of the income distribution most significantly in percentage terms. A similar pattern is observed for non-labour income, which tend to favour the wealthiest Canadians, and particularly since the 2008-2009 Financial Crisis. Time series evidence highlights a growing connection between policy surprises and real asset prices, with a more modest impact on unemployment. Altogether, these essays address crucial issues related to monetary policy, emphasizing the importance of evidence-based analysis and objective quantitative research in evaluating the effectiveness and consequences of central bank communications and policies.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.