Chungeun Koo, Jong-sung You, Sam Ock Park, Junyoung Seo, Seongjun Ahn
This paper describes the current state of the preliminary KIRL data deployment and discusses its possible applications and limitations. This data is constructed from the National Health Insurance Service (NHIS), the Survey of Household Finances and Living Conditions (Statistics Korea), and the Social Security Insurance Service (SSIS). The KIRL also constructed the combination data of pseudonymized information from NHIS and SSIS, and NHIS and Statistics Korea.
The NHIS data cover a national sample of all lawful residents in Korea, and we preliminarily built the following inequality-related indicators at the Si/Gun/Gu unit (city and district level) with about 16% of the sample data which covers 2003-2020: Theil index of income and property tax base, the total income and wage income share of top 10% and 1%, mean and standard deviation of income and total property tax base, the percentage of residents with national top 10% of total/wage income in each region. Other statistics such as education, the share of experts, and the percentage of workers in manufacturing from Statics Korea are merged into this dataset.
It should be noted though that the income data from the NHIS are not appropriate for the time series analysis because the income identification rate of the national tax service is changing (getting better). Therefore, we will improve the top 1% and 10% share using the household income data from the national account in the next step, instead of the NHIS dataset. The household income data from the national account are measured by the levels of the nation and subdivisions (Sido and Si/Gun/Gu) in Korea.
The data from the SSIS are data that only has information about beneficiaries of social security programs, and we have constructed data for 2012-2021 on the timing and the amount of benefits, as well as various types of income and assets for recipients of some major programs such as the National Basic Livelihood Security System. The SSIS data is combined with the NHIS data, which covers 14% samples of the entire population, in an outer single join method, which also provides data to compare social security recipients and non-recipients.
A project we’ve worked on is a project combining pseudonymized information from the NHIS and the Survey of Household Finances and Living Conditions conducted by Statistics Korea. This project aims to add information on household characteristics and detailed income and consumption-related information to the data involving the use of medical services. These combined data enable researchers to analyze, for example, the relationship between the use of medical services and the policy holders’ economic status.
The KIRL data have been used for statistical analysis for research which requires detailed income and property information on the fine unit or narrow geographies. An example is the study currently ongoing data-combination project, which is a part of the Jeonbuk research project. We analyzed the income distribution, inequality in income and property, and the population movement by income and property for this project. Other studies on policy effect analysis or other related studies.
수정
삭제
글쓰기
|