Has Labor Instability of Middle-Aged Men Been Universalized in South Korea? : Comparison of Work Trajectories between Cohorts and Educational Levels

Background and Purpose:

In South Korea, the increase in labor instability has been a serious social problem. Among them, one of the distinctive phenomena of Korea is the sudden destabilization of labor, both in employment and income, that occurs in the middle ages of the 40s and 50s. Since the Korean financial crisis in the late 1990s, many people retire from their lifetime jobs involuntarily, and these retirements take place at much earlier age than in other countries. Further, after retiring, many move to low-paid precarious jobs or start self-employment because of the low security of old income. Nevertheless, the labor instability occurs in middle age has been regarded only as a banal fact that rather prevented an in-depth analysis of the issue, thereby making effective policy approaches difficult. Based on the ‘generalization of risk’ discourse of German sociologist Ulrich Beck, this study examines whether Korean middle-aged men's labor instability is gradually increasing through generational comparisons, and analyzes whether vulnerability to the instability varies according to one’s educational level.

Methods: Data and samples: This study uses the 1998-2017 Korea Labour and Income Panel Survey (KLIPS) study sample, twenty years of personal data and 2017 work history data. The sample includes 1,969 male individuals, with Cohort 1 who was born between 1939~1948 (n=910) and Cohort 2 who born between 1949~1958 (n=1,059).

Measures: First, I measure instabiliy level from 0 to 4 by combining employment status and wage level each year. By linking each year's instability level, individual long-term data for 15 years, from 45 to 59, is constructed. Secondly, to examine the work trajectories and cluster similar trajectories, a group-based trajectory analysis is conducted. Group-based trajectory modeling(GBTM) is a finite mixture modeling application that uses trajectory groups as a statistical device for approximating unknown trajectories across population members. Thirdly, multinomial logistic regression analysis was performed to test whether cohort and educational levels have a significant effect on conditional probability of belonging to specific work trajectories.

Results:

Comparing the distinctive patterns of each work trajectories (total six trajectory groups) between two cohorts in various aspects, it turned out that instability in middle age became more stratified than universalized. In other words, the birth cohort born from 1949 to 1958 is more likely to belong to a stable trajectory than cohort who were born between 1939 and 1948, as well as the power of educational attainment to lessen ones’ working life instability has been greater.

Conclusions and Implications:

These findings indicate that even under the powerful influence of globalization, working life instability in middle age has decreased in Cohort 2 than Cohort 1, and this risk is differently distributed among individuals by one’s educational background. This study shows one of the important natures of the risk experienced by middle-aged people in Korea. In addition, the result suggests that specialized policies for low-education groups who have maintained a relatively unstable working histories are essential, as most policies currently target those with relatively stable jobs.