Tax Incidence, Progressivity, and Inequality in Canada

Knowledge about the distribution of the burden of taxes - as measured by their progressivity and their inequality impacts - is crucial for tax policy choices. Yet actual practice for both the formulation and assessment of tax policy does not draw on much of the best research knowledge. This study offers the first comprehensive critical survey of the field for Canada in nearly 20 years, a period of wide-ranging refinement and extension of research methods. We group the existing field of research into three principal genres. Inequality (INEQ) studies measure the inequality reduction from taxes borne directly by individuals, principally the personal income tax. Computable general equilibrium (CGE) studies examine the distribution of lifetime utility burdens of stylized taxes using complex mathematical economic models. Fiscal incidence (FINC) studies compute the pattern of progressivity or regressivity for each tax and the entire tax system using microsimulation methods. We assess the relative strengths and weaknesses of each type of study. We offer a compact overview of the measures of inequality and tax progressivity used in the empirical literature, followed by a review of the methodological issues that arise in measuring economic well-being for tax distributional analysis. We provide some evidence on the relative equalizing effects of transfers and personal taxes in Canada. Then we examine the comparative findings and methods of representative studies of each type, with emphasis on studies that include Canadian taxes and with a focus on the underlying assumptions about tax incidence. Transfers are found to be more important than income taxes in reducing inequality of Canadian incomes for all periods since 1971 and for most types of households (except those with high per capita incomes). INEQ studies find that Canada's ranking in inequality reduction from personal income taxes is intermediate among countries and dependent upon the measure of inequality; some studies find Canadian personal taxes to be less equalizing than the US counterparts. CGE studies have been developed most for analysis of the US tax system, with little comparable available for the Canadian tax system though the US results are suggestive of the Canadian situation. Based on lifetime income groups in the long-run equilibrium, only the personal income tax is found to be strongly progressive. Payroll taxes are strongly regressive; sales, excise, and property taxes are significantly regressive except for the top two deciles of lifetime incomes; and even the corporate income tax is somewhat regressive over lifetime income groups except for the top decile. All taxes taken together are found to be roughly proportional for the bottom nine deciles and highly progressive for the top decile. FINC studies using annual data, which have been most frequently applied for Canadian taxes, find either slight or substantial progressivity for the tax system overall; a lifetime study finds somewhat less progressivity than comparable results based on annual data. As with CGE findings, the FINC studies assign a key role to personal income taxes in any net progressivity of the total tax system, given the regressivity of many other tax types. Our analysis gives special attention to the economic basis for assumptions about the incidence of the major tax types used in the three types of studies. Tax incidence, and the possible shifting of tax burdens from the taxpayer to other parties, plays a critical role in analysis of the distribution of the tax burden. Multifaceted theoretical and empirical research casts doubt on the standard assumption that the personal income tax is borne fully by individual taxpayers. This evidence suggests that personal taxes on higher earners are at least partially shifted onto other parties, thus reducing the effective progressivity of the tax. Given the key role of personal tax progressivity in many studies' findings of overall tax progressivity, this issue warrants further research. If one were to use incidence assumptions more consistent with the cited evidence for the personal income tax, even these mildly progressive findings might be overturned. Most of the Canadian studies reviewed here are already quite dated in their periods covered and would benefit by updating to include the important tax policy changes since 1988. Also, the use of data sets permitting inferences about lifetime effects would permit better assessments of income-based versus consumption-based taxes. Still, priority in future research should be given to improved understanding of the incidence of personal taxes.

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.