The Marginal Cost of Redistribution: Lifetime and Annual Perspectives

Abstract

Approximately 40% of Federal government expenditures are spent on redistributing income from high income households to low income households [20, 293]. Changing income redistribution policies causes distortions which affect the efficiency of resource allocation. Costs caused by these distortions are an important consideration in determining redistribution policies and practices. Previous studies, such as Browning and Johnson [5] looked at the marginal cost of redistribution using annual income data. However, as is shown in this article, using annual rather than lifetime income data greatly understates the marginal cost of redistribution. The time period is an important consideration when calculating the marginal cost of redistribution. Over a lifetime a person may be a young student expecting a higher future income; be retired having had a higher income in the past; be changing jobs or recently divorced. Lifetime income includes these age and transitory differences in earnings and thus if we are concerned about the cost of decreasing inequality, lifetime income is more accurate. Also the relevant time period partly depends on the capability of households to even out income fluctuations by spreading income from good to bad times. Upper income groups find it easier to spread income by borrowing or waiting until income rises again while lower income groups find it more difficult to borrow or wait. If we are measuring the number of people in poverty and poor people find it difficult to spread income from good to bad periods, the relevant period of time may be about a week. But using a week for upper income groups would be misleading as many would appear poor but not face the difficulties of the poor. If most people can average their income, then the annual distribution of income is more appropriate than weekly. However, because of the above mentioned age and transitory differences in income, annual income may also be misleading. People are affected differently in different years by the tax-transfer system, in some years a person may receive transfers and in others pay taxes. Since households live their lives rather than a single year under the tax-transfer system, well-being is more accurately described on a lifetime basis. Lifetime income gives a more accurate picture of households' chances in life and of the degree of economic inequality. In this article, the annual and lifetime marginal costs of redistribution are calculated by a simulation that uses a lifetime dynamic labor supply model. A linear income tax type of redistri-