The Impact of Pensions on the Risk of Asset Poverty Among Older Adults In South Korea: Noncontributory Versus Contributory Pensions Using Inverse Probability Weighting

Background : Assets serve as a fundamental resource for ensuring well-being in later life. Assets safeguard households against income shortfalls and help maintain living standards and expand future opportunities. Following retirement, when stable labor income declines, assets become a critical substitute for maintaining livelihoods. In addition, pensions constitute a key pillar of financial security in later life, not only by providing a steady income stream, but also by preventing the depletion of assets among older adults. However, there remains a significant gap in the literature regarding the extent to which South Korea’s dual pillar pension system serves as a buffer against asset poverty among the elderly. This study investigates the effects of two distinctive pension programs—the contributory National Pension (NP) and the non-contributory Basic Pension(BP) for those in the bottom 70% of the income distribution—on the risk of asset poverty.

Method : Data for this study were drawn from the ten waves of the Korea Welfare Panel Study, spanning from Wave 11(2016) through Wave 20 (2025). The study sample consisted of elderly individuals who were not in asset poverty at the baseline of the survey (N = 1,951). The dependent variable was the length of time until the transition into asset poverty. Asset poverty was defined as possessing net wealth below the threshold of six months of minimum living expenses. Pension receipt was measured with two binary variables: the first indicated the receipt of BP exclusively, while the second identified the receipt of NP exclusively. Meanwhile, elderly individuals receiving both pensions were excluded from this study to ensure a clear distinction between the effects of each pension. The model adjusts for several key covariates, including gender, age, marital status, and household size, educational attainment, residential area, income, and baseline assets. While gender is specified as time-invariant covariate, all other variables are incorporated as time-varing. This study conducted the Cox Proportional Hazards models with inverse probability weighting (IPW) to adjust for the pre-existing differences between pension recipients and non-recipients.

Results : Among the sample, approximately 20% of older adults experienced asset poverty annually in Korea. However, the Kaplan-Meier analysis reveals that approximately 80% of the BP exclusively experienced asset poverty during the study period, while this rate was notably lower at around 50% for the NP exclusively. The primary analysis reveals that older adults receiving BP exclusively showed a lower risk of experiencing asset poverty (HR = .596, p<.001), compared to non-recipients. However, no significant association was found for those receiving NP exclusively.

Conclusion : The study found no significant differences in transitions into asset poverty between the non-recipients and NP recipients. However, BP recipients exhibited a lower likelihood of transitioning into asset poverty, confirming the effectiveness of BP. This difference likely reflects that recipients of BP, who are generally more vulnerable, benefit from transfers that help compensate for insufficient income, whereas NP benefits are not sufficient to offset income inadequacy among their recipients. Therefore, enhancing the adequacy of NP benefits is essential for reducing economic vulnerability among Korea's aging population.