|
||||||||||||||||||||||
Income Inequality Health Insurance Ridge Regression Graduate MACRO INTRO intro lecture notes email me |
Income Inequality Download the paper (PDF) Of all the papers I've written, this one is my favorite. It derives a very simple theory of the relationship between development and inequality from Solow's (1956) classic model of economic growth. To put the paper in context, in the United States, income inequality has been rising for decades:
And the effects of that increase in inequality have been acutely felt in the past few years as average income has risen, while median income has fallen:
In fact, the poorest 60 (sixty!) percent of households saw their real incomes decline between 1999 and 2006. How are we to understand these developments? The model developed in the paper asserts that tax policy may explain some of the differences in income inequality among states. According to the model, steady-state income inequality depends entirely on household saving rates. Relatively high-saving households accumulate more capital than relatively low-saving households and therefore have higher steady-state income. Households are assumed to save out of disposable income, so raising tax rates reduces household saving rates and causes the economy to converge to a lower steady-state level of average income. Different forms of taxation have different effects on inequality however. Taxation of labor income causes a larger percentage decline in the saving rates of relatively low-saving households, thus increasing steady-state income inequality. By contrast, taxation of capital income has the opposite effect: it causes a greater percentage decrease in the saving rates of relatively high savers, thus reducing steady-state income inequality. Data from the March CPS and the NBER's TAXSIM provides some empirical support for the model. Higher wage taxes are associated with a greater degree of inequality, even when the regression model includes educational inequality as a control for the fact that a tax on wages taxes the returns to both labor and human capital. Of course, taxation of wage income taxes the returns to both labor and human capital, but given the fact that most people invest in their educations before entering the labor force, it seems likely that the marginal effect on labor is much stronger than the marginal effect on human capital. A high degree of collinearity among the tax rates and the various forms of capital income taxation prevent us from discerning the empirical sign of taxes on capital income, but taxes on long-term gains, which exhibit the lowest degree of correlation with the other tax rates, are associated with lower inequality (as predicted by the theoretical model). Although the model and empirical evidence strongly suggest that tax policy has a role to play in shaping the distribution of income, I do not believe that tinkering with the tax code should be the only response to the increase in income inequality. The whole point of the paper is that if we can encourage low income people to save more, we can bring the economy to a higher level of economic development and reduce the degree of income inequality. Tax policy should be changed to encourage low-income families to save more, but that should not be the only policy response. We should also look deeper to find the fundamental reasons why some households save more than others and try to find ways to address the causes of low saving. For example, if we interpret capital as human capital and if human capital is acquired through education, then reducing high school dropout rates will raise the "saving rates" of low-income households and reduce steady-state income inequality. In future research, I plan to identify the characteristics of people who move up the income ladder (i.e. people who increase their saving rates) and find out if there's a way to "give" those characteristics to people who otherwise would not move up the income ladder. References R. M. Solow. "A Contribution to the Theory of Economic Growth." Quarterly Journal of Economics, 70(1):65–94, Feb. 1956. |