The Measurement and Trend of Inequality: Comment
批评了Paglin提出的修正基尼系数的方法，指出其错误分解导致低估美国收入集中度，并建议不要使用该指标。
In a recent article in this Review, Morton Paglin proposes a measure of concentration which he claims will correct the Gini ratio for overstatement in populations where each individual may be expected to vary his income or wealth over his life cycle. The Paglin measure suggests that the Gini ratio overstates income concentration in the United States by a third and does not reveal a 23 percent decline in concentration which has occurred between 1947 and 1972. These results are based on an improper decomposition of the Gini ratio. In this note I decompose the Gini ratio to derive the Paglin measure. It is shown to correct for neither demographic nor life cycle income effects and to understate intracohort income or wealth inequality.' If properly normalized, the Paglin measure shows that the Gini overstates U.S. income concentration by 7 percent or less, and indicates a decline in U.S. interfamily income concentration of less than 12 percent. Four percentage points of this decline are the result of demographic change, while less than 7 points (comparable to the 5 percent decline of the Gini) are the result of decreased income concentration within cohorts of family heads. Decomposition further reveals major changes in the division of income among cohorts and in concentration of income within cohorts. These changes compensate each other in the Paglin measure, the use of which as a measure of income or wealth concentration is not recommended. The Gini concentration ratio