Simulation Analysis of Tax Reform
Multi-Period Overlapping-Generations Growth Model
The tax reform whereby consumption tax is raised and labor income tax is reduced has been intensively investigated by using simulation analysis.
The first classical study is a paper by Summers (1981). In this paper, the author used an overlapping-generations growth model whereby each generation lives for 55 years and works for 40 years. Then, each generation retires for the last 15 years. In this regard, Summers investigated the quantitative effect of tax reform.
A household chooses its optimal consumption and saving for 55 years based on present value budget constraint. The population grows exogenously, technological growth is included, and labor supply is exogenous. GDP is produced by two inputs, labor supply and capital in the aggregate production function. Output is distributed to labor and capital owners according to marginal productivity. The government collects a given amount of taxes by imposing a labor income tax, a capital income tax, and a consumption tax.
According to Summers’s analysis, the most desirable tax to maximize long-run welfare is a consumption tax. The second is a labor income tax, and the last is a capital income tax. The difference between a consumption tax and labor income tax originates from the timing effect. As explained previously, a consumption tax stimulates saving and capital accumulation, enhancing long-run welfare. The difference between a labor income tax and a capital income tax is with respect to the interest elasticity of saving. If the elasticity of saving is large, capital income tax reduces capital accumulation to a significant extent.
The reason why elasticity is large is because of the human capital effect, as explained in Chap. 8; namely, an increase in the after-tax interest rate reduces the present value of future labor income, reducing the effective income. This depresses present consumption and hence stimulates saving. Considering this effect, even if the conventional substitution effect is not large, the interest elasticity of saving could be significantly large. If so, a reduction of the capital income tax rate stimulates saving and capital accumulation to a significant extent by reducing the after-tax rate of return on saving.
To sum up, there are two reasons why the consumption tax prevails over the labor income tax in improving welfare under the simulated reform in Summers’s model. First, the capital intensity of production is higher under the consumption tax reform. This raises the wage level and hence consumption above the increase due to the labor income tax reform. This is because the economy moves closer to the golden rule level of capital under the consumption tax reform than under the labor income tax reform.
This situation mainly occurs because under a labor income tax, the taxpayer pays the bulk of her or his taxes when she or he is working; whereas under a consumption tax, the tax liability is more evenly spread over the life cycle. Thus, under a consumption tax, the taxpayer has to save more when working in order to pay her or his future tax than she or he does under the wage tax; thus, saving and the capital intensity of production are higher, as explained above.
Second, the present value of the taxpayer’s tax liability is lower under the consumption tax reform than under the labor income tax reform. This occurs because those who are alive at the time of the transition experience a heavier tax burden than those who are living in the new, post-reform steady state. However, steady-state welfare calculations may be a poor indicator of the true costs and benefits of a tax reform, as noted by Summers. Further, the transition may be rapid; hence, only a few generations may be less prosperous under the proposed reform.