مالیات مترقی در یک مدل سلسله از سرمایه انسانی

We develop a quantitative theory of economic inequality to investigate the effects of replacing the current U.S. progressive income tax system with a proportional one. The cross-sectional implications of the theory are used to discipline the assessment of the effects of tax policy and circumvent the lack of conclusive micro-evidence on the parameterization of the human capital production technology. We find that the elimination of progressive taxation increases steady state level of output by 12.6%, capital by 21.8%, and consumption by 13.2%. Moreover, it increases economic inequality and its persistence across generations.

Despite the fact that income is taxed progressively in the U.S., most quantitative analyses of U.S. income taxation focus on proportional income taxation. Scholars in public finance are thus faced with the following question: does the progressivity of the tax code matter for evaluating the aggregate and distributive effects of U.S. income taxation? In this paper, we develop a quantitative theory of economic inequality in order to evaluate the effects of replacing the progressive income tax system in the U.S. economy with a proportional one. Building a theory of inequality is the first necessary step for assessing the consequences of progressive income taxation: since marginal tax rates increase with income under a progressive tax system, income taxation has a differential effect on individuals across the income distribution and, therefore, the distribution of income matters for the impact of taxation on the economy. Building a theory of economic inequality is also important for evaluating the effects of income taxation on human capital accumulation. While economists consider human capital as a crucial component of aggregate wealth, they have conflicting views about how human capital accumulation is affected by taxation. While King and Rebelo (1990) and Rebelo (1991) have found large negative effects of taxation on human capital accumulation, other studies have found negligible effects; see, for instance, Heckman (1976) and Davies and Whalley (1991). In an influential paper, Trostel (1993) showed that the consequences of income taxation hinge crucially on the specification of the human capital production technology. When time is the only input in the education technology, the reduction of the net wage due to an increase in the income tax rate equally affects the benefits and costs of human capital investment; in this case, income taxation does not affect human capital accumulation. However, when goods are an input to the human capital production, income taxation has an important effect on human capital because the cost of these goods is not reduced by income taxation. The conflict in the views about the impact of taxation on human capital accumulation has remained due to the lack of conclusive micro-evidence on the parameters of the human capital technology. In light of these difficulties, our paper provides a novel approach to studying the impact of income taxation on human capital accumulation. We build a model with heterogeneous individuals, which allows the cross-sectional implications of the theory to be used in parameterizing the human capital technology. Motivated by the empirical studies of Neal and Johnson (1996) and Keane and Wolpin (1997), we focus on investments that take place ‘early’ in the life of an individual and formulate a model of parental investments in the human capital of their children. In the model, individual labor productivity is jointly determined by education and an uninsurable market luck shock (which is iid over time and individuals). The model's main novelty relative to the previous work in the area is the inclusion of a production technology for human capital which takes expenditures and (quality-adjusted) parental time as inputs. Because the relative importance of these inputs determines how much of the cross-sectional earnings inequality is transmitted to the next generation, the intergenerational correlation of earnings is used to pin down the shares of time and expenditure inputs in the human capital production function. We find that modeling parental investments in human capital substantially increases the aggregate effects of replacing progressive income taxation with a proportional one. For the benchmark economy, this change in tax policy increases the steady state levels of output by 12.6%, capital by 21.8%, and consumption by 13.2%, while for an economy without investments in human capital (with exogenous labor productivity) the increase in the same variables is less than a half.1 Two main channels explain why progressive taxation has a much bigger impact in the economy with parental investments. First, our calibration implies a large share of expenditures in the human capital production function, which means that income taxation affects disproportionately the benefits and costs of human capital accumulation. Second, a progressive tax system treats asymmetrically the benefits and costs of time inputs in human capital production. When the marginal tax rate depends on income, the tax rate applied to the cost of time inputs is not, in general, equal to the tax rate applied to the return on human capital investments. As parents increase the time spent educating their children, their earnings fall thus lowering their marginal tax rate. At the same time, higher human capital of the children increases children's earnings thereby moving them to a higher marginal tax rate. This increases the wedge between the tax rates applied to the costs (foregone parental earnings) and benefits (higher earnings of offspring), creating additional disincentives for human capital accumulation. This effect was not examined by Trostel (1993) because he studied a flat income tax in a representative agent environment. The elimination of progressive income taxation leads to an important increase in economic inequality: the Gini coefficients increase from 0.289 to 0.304 for earnings, from 0.457 to 0.519 for after-tax income, and from 0.781 to 0.843 for wealth. Moreover, a version of the model with exogenous labor productivity generates slightly less pronounced inequality effects. We find that the effects of progressive taxation on economic mobility, as measured by the intergenerational correlation of earnings, depend crucially on the specification of the human capital technology, a result that is novel in the literature. In particular, progressive taxation can either increase or decrease economic mobility depending on the importance of the time and expenditure inputs in the accumulation of human capital. In the presence of time inputs, progressive taxation creates strong incentives for parents with high market luck to invest in the human capital of their children. Because parents with above-average market luck expect their children to have lower market luck, they also expect their children to face a lower marginal tax rate than themselves. They thus face strong incentives to spend time educating their children, which leads to a higher intergenerational correlation of earnings under progressive than under proportional income taxation. On the other hand, a progressive tax system can enhance economic mobility in the presence of expenditure inputs. Nondeductibility of expenditures discourages investments in human capital accumulation. When income taxes are progressive, this effect is more important among parents with high earnings because they face a high-marginal tax rate on their investments. To sum up, the imperfect deductibility of time inputs and the lack of deductibility for expenditures under a progressive tax system have opposite effects on earnings mobility. Our experiments show that the second effect is more important: eliminating progressive taxation reduces economic mobility.

We develop a quantitative theory of economic inequality to investigate the effects of replacing the current U.S. progressive income tax system with a proportional one. In particular, we use the cross-sectional restrictions of the theory—such as the intergenerational correlation of life-time earnings, the variance of life-time earnings and its decomposition into endogenous and exogenous components—to jointly pin down the parameters governing the human capital technology and the random earnings process. We find that modeling parental investments in human capital substantially increases the aggregate effects of replacing progressive income taxation with a proportional one. For the benchmark economy, this change in tax policy increases the steady state levels of output by 12.6%, capital by 21.8%, and consumption by 13.2%, while for an economy without investments in human capital (with exogenous labor productivity) the increase in the same variables is less than a half. Two channels account for the difference. First, our calibration implies a large share of expenditures in the human capital production function, which means that income taxation affects disproportionately the benefits and costs of human capital accumulation. Second, a progressive tax system treats asymmetrically the benefits and costs of time inputs in human capital production. When the marginal tax rate depends on income, the tax rate applied to the cost of time inputs is not, in general, equal to the tax rate applied to the return on human capital investments. Moreover, we find that the nondeductibility of expenditures and the imperfect deductibility of time inputs under progressive income taxation have, in isolation, important effects on earnings mobility. Because these effects work in opposite directions, earnings mobility decreases only by a modest amount with the elimination of progressive taxation. Note that our analysis abstracted from the labor-leisure choice. Introducing this feature into the model economy would increase the aggregate impact of progressive taxation since labor supply distortions further discourage human capital accumulation. This non-trivial extension is left for future research.