Poverty Transitions for the Elderly

TitlePoverty Transitions for the Elderly
Publication TypeThesis
Year of Publication2009
AuthorsLee, Y
Number of PagesPh.D.
Date Published2009
UniversityThe Ohio State University
CityUnited States -- Ohio
Thesis Type3340373
KeywordsHealthcare, Methodology

The purpose of this study is to examine the likelihood of poverty transition and the effects of different events on poor elderly. This study is important because the aging population has grown rapidly, and elderly poverty involves many complex relationships across an individual's life span. The contributions of this study are to improve the understanding of elderly poverty and to provide considerable policy implications for elderly demographic changes in the future. The data for this study are from the years 1992-2006 of the Health and Retirement Study (HRS) and the sample consists of 30,405 elderly individuals from eight waves. To examine the incidence and dynamics of elderly poverty, poverty-rate decomposition and the poverty exit (re-entry) hazard rate based on Kaplan-Meier product-limit estimates are used. Entry into and exit from elderly poverty models are separately used to estimate the conditional relationship between poverty transition and multiple trigger events as well as various covariates using the discrete-time hazard model. These multivariate techniques show a more realistic picture of elderly poverty transition in terms of providing a preliminary explanation of the unobserved heterogeneity of the elderly poverty. The major findings are: (1) the poverty rates for the HRS data fluctuated considerably during the 1990s, but the rates have had little turnover and have been relatively stable over time during the 2000s; (2) in terms of poverty entry and exit rates, the exit rate was decreased during the 1990s, but the rate was increased during the 2000s while the entry rate fell somewhat during the 1990s and rose somewhat during the 2000s; (3) the poverty rate for the HRS cohort individuals in a given year by cross-sectional data is relatively low, and the duration of the poverty spell is also relatively short (a fifth of the HRS cohort individuals had at least one poverty spell); (4) as the length of the poverty spell increased, the probability of poverty exit decreased; (5) as the non-poverty duration increases, the poverty re-entry rates are constant at around 10 percent; (6) retirement and a negative change in health condition both have significant effects on elderly poverty entry, while retirement, increase in total wealth, and becoming insured from any government health program all have significant effects on elderly poverty exit; and (7) life history variables, such as total years of work and length of marriage have significant effects on both elderly poverty entry and exit. Results from the hazard rates (exit and re-entry rate) imply that a person who falls into poverty during his or her elderly years is highly likely to remain poor because the exit probabilities fall as the length of the poverty spell increases. In addition, the results of constant re-entry rates infer that the elderly population is exposed to the risk of falling to an income that puts them below the poverty line. Results from multivariate analysis suggest that retirement has an important role in elderly poverty transition and a negative change in health condition also has positive impact on elderly poverty entry. Thus, effective income support programs and social policies for the elderly help to prevent elderly individuals from becoming poor.

Endnote Keywords


Endnote ID


Short TitlePoverty Transitions for the Elderly
Citation Key6405