서강대학교

초록/요약

This study investigates the characteristics that contribute to elderly poverty, mainly focusing on individuals’ lifetime work experience. It adopts the heterogeneous relative poverty line which differs by gender, province of residence and time of survey. It calculates the work experience and demographic variables using the Korean Labor and Income Panel Study’s (KLIPS) survey data for 2006, 2009, 2012 and 2015. The objective of the study is to estimate the incidence of poverty and its severity amongst elderly and explain the variations in poverty in relation to individual characteristics and lifetime work experience. It calculates the head count, poverty gap and the poverty severity indices as poverty measures. The poverty measures in this study are based on the monetary dimensions of well-being -- income and consumption. Consumption is smoother and more closely related to individual needs. The methodology used in this paper is the logit model to explain incidence of poverty and the Heckman sample selection model to analyze the depth and severity of poverty. The results of the Heckit model show evidence of a significant selection bias in all the poverty models based on income, but not on the consumption. The empirical results in both income and consumption models show that an increase in the total work years lessens the incidence of poverty and a decrease in the gap years downsizes the probability of being poor. High-income occupation and labor market participation greatly decrease the incidence of poverty. Most of the work relevant variables become insignificant in the poverty gap and severity models of consumption while both work years and gap years are significant in the income model. The number of jobs representing turnover rate significantly increases the probability of being impoverished only in the consumption model.