Estimation du taux de chômage naturel régional : le cas des régions administratives du Québec
|Abstract:||Regional unemployment in the province of Quebec is characterized by a significant heterogeneity both in the levels across the regions and how they evolve. Between March 1997 and December 2018, for example, the mean unemployment rate was about 6.3% in Capitale-Nationale and of 10.0% in Saguenay-Lac-Saint-Jean; an important gap of almost four percentage points. Because those disparities are inefficient (Taylor, 1996) and because they exacerbate income inequalities (Macphail, 2000), understanding their determinants is essential. Previous works on the determinants of unemployment point to the influence of fiscal legislation and labourmarket institutions to explain unemployment rates disparities amongst the countries. Therefore, those factors being homogeneous for regions in the same province, they cannot explain this diversity between Quebec’s administrative regions. Based on Friedman’s (1968) definition of structural unemployment, we develop a model that divides the observed regional unemployment rates in a natural (or structural) and a cyclical component which allows for the integration of new determinants like regional productivity and industrial composition. Using panel data on Quebec’s administrative regions from March 1997 to December 2018, we run a fixed effects ordinary least square (OLS) estimation with an SCC error term robust to correlation and cross-correlation plus a fixed effects general least square (GLS) estimation integrating an AR(1) correlated error term. Results show that regions’ productivity and industrial composition both have a significant effect on regional unemployment. Finally, we use our results to calculate the natural unemployment rate and its evolution for each region andbriefly discuss its evaluation.|
|Document Type:||Mémoire de maîtrise|
|Open Access Date:||2 August 2021|
|Collection:||Thèses et mémoires|
All documents in CorpusUL are protected by Copyright Act of Canada.