Essais sur les politiques macroéconomiques

This thesis undertakes a comprehensive analysis of the influence of fiscal and monetary policies on the economy. The first two chapters primarily involve empirical investigations utilizing Vector Autoregressive (VAR) models. In the aftermath of the Great Recession, economists have dedicated considerable attention to the efficacy of these policies in reviving economic activity. The first two chapter of this thesis makes a notable contribution to the ongoing discourse by delving into the following specific aspects: the impact of monetary policy communications (“forward guidance”) and the evaluation of the fiscal multiplier. In the third chapter, I delve into the examination of how households’ heterogeneity affects the transmission of monetary policy within a Tractable Heterogeneous Agents New Keynesian (THANK) model. The primary objective of the first chapter is to investigate the effect of US forward guidance and unanticipated monetary policy easing on four US bilateral nominal exchange rates (US dollar against Canadian dollar, British pound, Euro, and Japanese yen), net exports, and real GDP. The underlying concept developed in the MundellFleming model is that a domestic country’ currency depreciation in response to monetary easing prompts a shift in expenditures towards domestic goods, thereby enhancing the nation’s trade competitiveness, net exports, and real GDP. This study introduces a novel element in the macroeconomic literature by addressing the empirical existence of this expenditure-switching effect triggered by forward guidance easing, an aspect hitherto unexplored in empirical macroeconomic literature. I commence by examining a Two-Country New Keynesian model’s response to unanticipated monetary policy and forward guidance easing concerning exchange rates, real GDP, and other macroeconomic aggregates. Subsequently, I apply the observed sign of the responses in the model to their empirical counterparts to disentangle the empirical effects of the two shocks using quarterly data from 1983 to 2019. The findings reveal that US forward guidance easing leads to a depreciation in exchange rates, ranging from 2 to 4 percent. This depreciation, in turn, initially impacts net exports negatively before it improves, as anticipated by economic theory. However, the credible sets of the responses (sets of possible empirical responses) encompass zero, highlighting a possibility that there is no effect of the depreciation on US’ net exports. The robustness of the results is established through the application of narrative sign restrictions and supplementary analysis. Given the major priority of identifying the repercussions of the shock on the macroeconomic aggregates, the second contribution of this chapter stems, among other things, from the demonstration that the use of the narrative approach optimizes the identification of the shock from macro quarterly data. The remaining supplementary findings confirm that (i) a VAR model augmented with interest rate forecasts contains sufficient information to identify forward guidance and unanticipated monetary shocks. Moreover, (ii) the nominal bilateral exchange rates exhibit a hump-shaped depreciation of two to four percent following a 25 basis point forward guidance easing, aligning with the Uncovered Interest Rate Parity condition. (iii) Both shocks account for between 7.3 percent and 27.9 percent of the exchange rate variance, with the forward guidance shock contributing to at least 50 percent of this decomposition. Finally, (iv) forecasters perceive the forward guidance shock as a deviation from the Taylor rule. The second chapter, a collaborative endeavor, focuses on evaluating the Canadian government spending multiplier. This research is particularly relevant due to the limited number of studies addressing the fiscal multiplier in the context of Canada. Furthermore, this work contributes to the ongoing debate regarding the state’s dependency and the value of the fiscal multiplier, which is generally considered to be state-dependent and falls between 0.5 and 1 for developed countries. For this study, we use a Vector Autoregressive (VAR) model employing quarterly Canadian data spanning from 1961 to 2019. The selection of variables for the VAR model is meticulously undertaken to accurately reflect the distinct characteristics of the Canadian economy, particularly its interactions with US business cycles. The impact of government spending shocks is identified through sign restrictions. The results highlight a multiplier greater than 1 for the Canadian economy. Additionally, the examination of state dependency of the multiplier reveals that the effect of government spending remains consistent during both periods of economic expansion and recession. Furthermore, neither net exports, real exchange rates, nor terms-of-trade exhibit significant responses to government spending shocks. The study also uncovers substantial and long-lasting effects of investment-specific and consumptionspecific government spending shocks on output. In the final chapter, the analysis extends a Tractable Heterogeneous Agent New Keynesian (THANK) model with housing sector to examine how disparities in access to investment and housing, particularly property ownership affordability, influence the transmission of monetary policy. Housing’s pivotal role as a primary consumption good and asset for households in the US and Canada is well established in macroeconomic literature. Fluctuations in housing prices and associated costs directly impact households’ wealth, influencing their decisions regarding consumption and savings. However, these effects are far from uniform across households. The implications of housing price fluctuations on consumption are contingent upon financial situations (income inequality) and access to property (housing inequality). Therefore, this chapter contributes to the literature and ongoing debate by proposing a Tractable Heterogeneous Agent New Keynesian (THANK) model to analyze the housing sector’s influence on monetary policy transmission. The model incorporates two categories of households: the wealthy households called ”Savers”, and the low-income group referred to as ”Hand-to-Mouth.” The former group allocates a portion of their income to both housing and capital investment, while the latter group fully spends their income on goods and housing rent, which reacts more than one to one to shifts in aggregate income. The results reveal that the combination of housing and capital investment increases the consumption multiplier in response to an interest rate cut, in comparison to a model exclusively based on capital investment or the New Keynesian model featuring a representative agent. Moreover, when considering income inequality and the risk of households transitioning between categories, the multiplier effect becomes even more pronounced. Additionally, when fiscal authorities redistribute different income types within the model, the predominant impact is observed on aggregate consumption and consumption inequality. Lastly, when the central bank targets a synthetic Consumer Price Index (CPI), encompassing both goods and housing rent, the effect of an interest rate cut on macroeconomic aggregates is diminished.

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.