Although it is well-known that in life cycle models of consumption and labour supply, capital income taxation affects the labour supply through the normal income effect, this interaction may be widespread. Based primarily on graphic tools, three channels through which capital income taxation may affect labour market behaviour, in a manner that it increases labour supply or work effort thereby reducing the distortionary effect caused by a wage tax, are identified: first, capital income taxes alter the lifetime labour supply when workers are constrained on hours of work; second, they affect labour supply in the case where consumers target a certain level of lifetime consumption; finally, they influence work effort in an efficiency wage model. Although the magnitude of these effects must be determined through careful empirical investigations, there are indications that it may be quite significant. Constraints on hours of work affect the majority of the employed labour force and surveys indicate that a large portion of well-to-do baby boomers target consumption rather than maximize lifetime earnings. The large cohort of baby boomers has also accumulated large financial assets for the purpose of purchasing additional leisure through earlier retirement. These financial assets increase the ratio of non-labour income to labour income and may affect the work effort of a large share of the working population.
Key Words: Tax reform, income tax, capital income taxation, labour income taxation, labour supply, work effort
A decade after the introduction of major tax reforms in Canada and the U.S., tax reform is again taking centre stage in both countries. The proposals advanced so far involve primarily a shift in the relative taxation of capital and labour income. Some proposals, such as the consumption-base flat tax developed by Hall and Rabushka (1995) for the U.S. and its versions for Canada proposed by Grubel (1995) and Fortin (1995), involve a shift in factor taxation from capital income to labour income. At the other end of the spectrum, proposals for a move to a comprehensive income base, such as that of Ruggeri and Vincent (1998), involve a reduction in the burden of taxation on labour income through across the board tax rate cuts financed largely by the elimination of tax preferences for that component of capital income which is currently tax-sheltered. The economic implications of the various tax reform proposals depend crucially on the response of those two factors to the changes in their relative levels of taxation. The debate on tax reform, therefore, can be viewed as a debate on the relative economic effects of different factor-specific taxes.
Because of the complexities in the process that generates the economic effects of tax changes, tax policy options have been increasingly analysed through the use of computable general equilibrium models (CGE). These models usually rest on a neoclassical foundation where firms maximize profits and individuals maximize utility by making temporal choices about work and leisure and intertemporal choices about present and future consumption. In this framework, individuals and firms have complete flexibility in their choices and can make marginal adjustments instantaneously. The implications of this approach is that factor specific taxes affect only the factors upon which they are imposed. A recent example of this approach is provided by Devereux and Love (1994) who used a two-- sector model of endogenous growth to measure the welfare effects of alternative tax regimes. As they point out explicitly "it is clear that the three types of taxes - consumption taxes, wage taxes and capital taxes have independent effects"(p. 515).
Life cycle models incorporate changes in economic behaviour during the various stages of an agent's lifespan, but usually treat the retirement decision as a given. Either they fix a certain age of retirement (Fougere and Merette, 1998) or assume that "individuals. …