The Incidence of the Corporate Income Tax on Wages: Evidence from Canadian Provinces

Résumé

Corporate income tax (CIT) incidence is an important and contentious issue in tax policy discussions. Much of the focus in the recent literature and in policy discussions concerns the allocation of the burden of the CIT between owners of capital and labour. Since income from capital tends to be concentrated with wealthier individuals, if the burden of the CIT falls largely on capital it increases the tax system’s progressivity. On the other hand, if the tax is borne mostly by labour through lower wages, the CIT is less progressive. Despite the importance of this issue in policy discussions, empirical evidence is quite limited and the results are mixed; there is a particular dearth of empirical research on the incidence of corporate taxes in a Canadian setting. According to theoretical open economy general equilibrium models, the burden of the CIT may partly, and possibly largely, fall on labour. In these models, an increase in the CIT reduces the return to capital, causing capital to leave the jurisdiction, which lowers the marginal product of labour and ultimately wages. Thus, the CIT can have a negative indirect effect on wages through its impact on labour productivity by way of its impact on capital. However, the magnitude of this effect depends critically on several modelling assumptions and parameter values related to the size of the country, the degree of capital mobility, the nature of competition in the output market, etc. An emerging empirical literature investigates the effects of CIT on wages by way of this indirect transmission mechanism. Empirical studies in this vein include Hassett and Mathur (2006, 2015) for a cross-section of countries; Desai, Foley and Hines (2007) and Felix (2007, 2009) for the U.S. They all find evidence in support of the relevance of the indirect channel using national aggregate data. Other studies, such as Carroll (2009) and Felix (2009) for the U.S., examine corporate tax incidence at the subnational level, and find that a substantial amount of the burden of the CIT falls on workers. However, a recent study by Clausing (2013) using OECD data casts some doubt on the relevance of the indirect channel and the empirical results of some of the above studies. Another strand of research has focused on an alternative channel whereby the CIT affects wages directly. In these models, firms earn economic rents due to imperfect competition and/or other market frictions. Firms and workers bargain over these rents, allowing workers to earn a premium over the value of their marginal product. If firms earn economic rents and bargain with workers over their distribution, then an increase in the corporate tax can affect wages directly by lowering the rents available for distribution. Again, the theoretical results can be sensitive to various modelling assumptions and the emphasis has been on empirically identifying the so-called direct effect. Studies in this vein include Felix and Hines (2009) for the U.S., Dwenger et al., (2011) and Fuest et al., (2015) for Germany and Arulampalam et al., (2012) for a cross-section of European countries. They tend to find some empirical support for the direct transmission mechanism, though estimates of the size of the effect vary. In this paper, we undertake one of the few empirical investigations of the incidence of the CIT on wages using Canadian data. We focus on the indirect transmission mechanism of corporate taxes on wages. To this end, we estimate wage and capital/labour ratio equations simultaneously, using a panel of provincial data from 1981 to 2014. In our most preferred specification, we estimate that the elasticity of the real hourly wage rate with respect to the statutory CIT rate at the provincial level is -0.107; thus, a one per cent increase in the provincial corporate income tax rate is associated with a 0.107 per cent reduction in the real hourly wage rate. A common approach to assessing the impact of an increase in the CIT on wages is to calculate the impact on aggregate wages of raising one more dollar in corporate tax revenue. We use this approach to calculate the incidence of the CIT in Canada’s 10 provinces implied by our elasticity estimate. Under the commonly employed assumption that the CIT base is insensitive to changes in the tax rate, our estimates suggest that a $1 increase in corporate tax revenues due to an increase in the provincial statutory CIT rate reduces aggregate wages by from 95 Canadian cents in Newfoundland and Labrador to C$1.74 in New Brunswick. In an innovation to this approach, when we account for the fact that the CIT base shrinks in response to an increase in the tax rate, our estimates are significantly higher, ranging from C$1.52 for Alberta to C$3.85 for Prince Edward Island. Our results provide empirical support in a Canadian setting for the indirect transmission mechanism highlighted in the open economy general equilibrium models of corporate tax incidence and suggest that workers bear a significant part of the corporate income tax liability in the form of lower wages. The empirical results are robust to various sensitivity checks and are within the range of values obtained in previous similar studies. The remainder of the paper is organized as follows. In section 2, we provide a brief overview of the literature on the incidence of the CIT. In section 3, we specify the empirical model and discuss the data. The empirical results are presented and discussed in section 4. Section 5 concludes.