최상위 소득 대체를 활용한 근로소득 불평등 측정

In this study, we employ top-corrected Gini coefficients to supplement the limits on which the survey data did not adequately reflect the top of the income distribution. We use both the National Tax Service of the 1000th-percentile recent income data and the Korea Labor and Income Panel Study(KLIPS) in 2013-2016. In the descriptive analysis, the ratio of the highest income in survey data was lower than that of the National Tax Service, implying that excluding the highest income earners could be biased in estimating earning inequality. Empirical analysis shows the comparison of Pareto Type I distribution using the National Tax Service data and the KLIPS income distribution. And we derive the result of top-corrected Gini coefficients. The analysis indicates that for the top 1% earning income level, the results of KLIPS are underestimated than those obtained from the Pareto Type I distribution. In detail, top-corrected Gini coefficients for 10%, 5%, and 1% of the highest income in the National Tax Service data shows lower inequality in 2016. As the replacement rate of the highest income higher from 1% to 10%, the degree of inequality is growing greater. In addition, comparing top-corrected, Household Income and Expenditure Survey and KLIPS Gini coefficients, the top-corrected Gini coefficient measures the worst income inequality. Therefore, this study highlights that the highest income may be understated in the survey data, and in order to accurately estimate earning inequality, it is necessary to utilize a revised Gini coefficient incorporating income tax data.