This presentation examines the poverty risk among Japanese youths using longitudinal data. According to the Organisation for Economic Co-operation and Development (OECD), the poverty rate in Japan (16.1%) is the 6th highest among the OECD member countries (OECD 2015). Since this figure is based on cross-sectional data, we cannot correctly interpret the number to determine whether all people face the 16.1% of poverty risk or the poverty is more likely to concentrate on that specific 16.1% of people (Breen and Moisio 2004). In addition, it is not clear whether high parental poverty risk extends to their descendants, or whether education is helpful in decreasing the poverty risk. In this presentation, the data from the Japanese Life Course Panel Survey (JLPS) is used. The survey was introduced in 2007, and is conducted annually. Samples (N=4,800) were randomly selected among Japanese men and women aged 20 to 40 years. Poverty can be defined on the basis of the annual household income, and this survey could reflect the changes in Japan’s household economy. Traditionally, the Japanese labor market is gender-segregated, and the social security and welfare system is family-centered. In other words, under the strong division of labor based on gender, people seem to be required to overcome life difficulties by themselves. Thus, I hypothesize that poverty risk would be more likely to concentrate on people from disadvantageous backgrounds. In addition, since the tuition fees for higher education in Japan are very high, education may not be helpful in overcoming poverty.