In Korea, the fertility rate has begun to decline at an unprecedented rate, and the elderly population growth has been outpacing the historical national average. Lowered fertility rates and improved health longevity are allowing the population to live longer. In fact, in 2000 the elderly aged 65 and over accounted for 7.2 percent of the total population; Korea has become an aging society. As a result of increased longevity, such demographic changes can greatly affect the economic security of overall elderly population.

In terms of their financial security, the elderly are the group most vulnerable to uncertainty, even though several income compensation programs have been established in order to transfer resources to the elderly. Many elderly could not recover from an income loss by working or from a large medical expense by borrowing against future labor earnings (Hurd, 1989). Most Korean elderly were less satisfied with their current income and they mostly depended on private transfers from family as a main income source. Furthermore, many elderly persons received financially inadequate income: in 2003, almost one-half of the elderly reported having no income, and in 2004, about 14 percent of the elderly received income from any pension (KNSO, 2005). As a result, the elderly who have unstable economic conditions have no option for supporting themselves at times of financial uncertainty such as an income loss or medical problems, as long as their family is responsible for helping them. However, the decline of the fertility rate, the likelihood of being in a nuclear family, and the lack of sufficient public assistance for the elderly have significantly accelerated economic vulnerability for the elderly in Korea.

Overall, most women of the elderly population are more likely to experience declines in income and the actual changes in economic well being. Many women face very difficult circumstances in life due to having experienced discrimination and economic vulnerability throughout their life (Choi, 2005; Choudhury & Leonesio, 1997; McDonald, 1997; Prus, 2000). Poverty was also disproportionately to females, divorced, abandoned, or widowed (Hardy & Hazelrigg, 1993). Women are more likely to be widowed due to their longer life expectancies and the propensity of men to marry younger; about half of women over the age of 65 were widows (U.S. Census Bureau, 2007). Older women without a spouse are more likely than men to face threats to their economic security. Those women who have not worked are dependent in their later years on their spouse’s lifetime earning record. In fact, it is the loss of a spouse and his economic resources that is associated with declines in the economic well-being of older women (Burkhauser, Butler, & Holden, 1991; Zick & Smith, 1991). As labor market involvement of women increases, older women contribute to employer-based retirement benefits for their own retirement income. However, more working men than working women (74 percent versus 69 percent) save for retirement, and women receive lower retirement benefits than men (U.S. Census Bureau, 2005). The risks of poverty for older women are directly related to certain life conditions, including low educational attainment, limited occupational opportunities, spousal death, dissolution of marriage, and so forth. Among various risk factors, some events later in the life course, in particular spousal death, have gender-differentiated impacts. Therefore, economic security in later life is the most pressing concern for elderly women with the rise of the aging population.

In the U.S., upon the loss of their spouse, women’s income can be dramatically reduced due to lower Social Security Benefits and loss of pension income. Three out of every four older poor individuals are women, with women being twice as likely to be living in poverty as men in the U.S. (Choudhury & Leonesio, 1997). The combination of the increasing divorce rate and women’s longer life expectancy greatly reduces the financial security for women in the older age group (Orel, Ford, & Brock, 2004).

In Korea, 41.2 percent of elderly women were in poverty, compared to 31.2 percent of male counterparts (Seok & Lim, 2007). In addition, Seok and Lim (2007) showed that the income level of elderly women was about less than one third of the average income of their male age-peers. Approximately 34 percent of elderly older women were the highest beneficiaries of the National Basic Livelihood Security System (NBLSS), which serves as a last resort for the lower income bracket (Kang & Kim, 2009).

Because of the high prevalence of poverty rate of elderly women and the continuous aging of the population in Korea, it is important to examine the economic status of elderly women in order to suggest policy implications for improving the later life of elderly women. The first part of this paper discusses the poverty rate of the Korean elderly women. In the actual analysis, absolute and relative poverty threshold were chosen as representing the economic status of this group. The second part of this paper discusses the income portfolio of the Korean elderly women. It is shown that the relative contribution of public and private transfers as main income sources to old-age income security is considered. The last part of this paper provides results of multivariate analyses to specify risk factors of elderly women in poverty.

In the past decades, the Korean government provided very limited social protection for the poor under the Livelihood Protection System (LPS). Although the LPS Act was amended in 1986, this program still failed to meet the demands for effective public assistance to the poor. According to the amendment, people aged 65 or older and children aged 17 or younger could receive LPS benefits if they did not have any family support or did not expect any financial help from their family. However, there are several problems with this program (Kim & Kim, 2004) first, an insufficient government budget for the LPS, about 0.2 percent of the GNP, was restricted in terms of the benefit amount; second, the selection procedure of recipients was unfair because it was conducted without a means test; and third, the level of benefit did not make adjustments for different-sized families using an equivalence scale.

The Korean financial crisis in 1997 triggered reform to the LPS since the number of the poor requiring urgent social support was rapidly rising. In 2000 the National Basic Livelihood Security System (NBLSS) replaced the LPS. Although the NBLSS did not have an age for eligibility, the elderly had a large program participation rate of 25.8 percent among total recipients (Kang & Kim, 2009). Therefore, the NBLSS contributes to the support of the elderly poor as a public assistance program. The NBLSS examines both financial criteria and family responsibility at the same time to determine eligibility. To be eligible for this program, individuals must have limited income and assets and must not have any family member liable to support them. In fact, a large number of the elderly poor are excluded from the NBLSS benefits due to the determinant of the family responsibility rule. Since the initial year of the implementation of NBLSS, the number of recipients has slightly expanded, but the recipient percentage of the elderly population has settled at around 8~9 percent.

Unlike people in developed economies, whose main source of income shifts from employment to pension income upon retirement, few Koreans have pensions. This is partly due to the country’s immature National Pension Scheme (NPS). The National Pension Act was enacted on January 1, 1988 for workplaces with ten or more employees. The compulsory coverage was gradually expanded, eventually becoming the national pension scheme for the public in 2006 (National Pension Service, 2006). The compulsory coverage was extended to workers at workplace with five or more employees in 1992, less than five employees in rural areas, farmers, and fisherman in 1995, and the urban self-employed and employer and employees at workplace with less than five workers in 1999. Other public pension schemes in Korea include Government Employees Pension implemented in 1960, Military Personnel Pension (1963), Private School Teacher’s Pension (1975), and Specially Designated Post Office Personnel Pension (1992). Because it is not a pay-as-you-go system, only limited numbers of the elderly are beneficiaries of the NPS. About 17.1 million people were enrolled in the NPS and 1.8 million people received the NPS benefits in 2005 (NPS, 2006). As a result of fiscal insolvency, the NPS income replacement rates began to reduced in 2008, with a planned total reduction to 40 percent by 2028. Despite this drop in income replacement percentage, the contribution rate will remain at nine percent.

In an effort to reduce elderly poverty and to complement the NPS, the Korean government introduced a means-tested income support program, the Basic Old Age Support Pension (BOASP), in January of 2008. About 60 percent of the elderly will receive five percent of the mean NPS, which is 84,000 won per month as the BOASP benefits. The BOASP program does not count private transfers for the means testing, whereas the NBLSS restricts its benefits to those who do not have any family members who are legally responsible or are financially unable to support them. Such eligibility criteria based on kinship assumes that informal support mechanisms will provide full old-age income support. However, with the weakening of these support mechanisms, the BOASP was introduced without consideration of kinship network. At the beginning of 2008, the Korean government amended the NPS and the BOASP, in order to (1) increase the BOASP benefits from five percent to 20 percent and expand the coverage from 60 percent to 100 percent of the elderly population, and (2) to reduce the NPS benefits from 40 percent to 25 percent, while maintaining the current contribution rate of nine percent.

Socio-Economic Status (SES) is an important attribute in many studies on economic status of the elderly. Although the consumption opportunities available to an individual or household are conceptually suitable for measuring economic status, income is the most widely used measure of economic status (Hurd, 1989). In order to examine the distribution of income, the poverty rate, the fraction of a population whose incomes fall below the official poverty threshold is often used to measure it. Thus, the socio-demographic characteristics of the elderly such as age, educational attainment, marital status, residence, wealth, and job history are important factors associated with elderly poverty (Choi, 2007; Choi & Ryu, 2003; Kang & Kim, 2009; McLaughlin & Jensen, 2000).

For the last 40 years, the improved economic security of the aged population in developed countries is linked to the maturation and expansion of social programs, such as in the United States’ Social Security, Supplementary Security Income, and Medicare and Medicaid programs (Barusch, 1994; Hurd, 1989). Since 2000, the elderly poverty rate in the U.S. has been below 11 percent, and that rate has remained stable over time (U.S. Census Bureau, 2007). Despite the well-developed social policies for the elderly poor and the rise of economic security in the elderly population over time, certain sub-population groups among the elderly still remain at high levels of poverty. Those groups include the very old and elderly women who head their own households or who live alone. Additionally, higher poverty rates among elderly women in the U.S. persist and tend to be more common than for men (Burkhauser & Holden, 1982; Minkler & Stone, 1985; Morgan, 2000). The poverty rate for older women in the U.S. is almost twice as high as that for men (Rupp, Strand, & Davies, 2003).

Women’s economic vulnerability in old age is caused by their lower lifetime earnings and insufficient retirement savings (McNamara, O’Grady-Leshane, & Wiliamson, 2003). Among non-elderly women, their low-paying jobs and occupational segregation further contributed to their poverty status (Heath & Kiler, 1992), and these economic circumstances could have a lasting impact on the economic outcomes of older women’s later life (Choudhury & Leonesio, 1997). O’Rand and Henretta (1982) found that among women, interrupted work histories-not working in midlife or entering the labor force after 35 years of age-were associated with significantly less retirement income than lifelong work histories. According to Choudhury and Leonesio (1997), older women who reported two years or less of work experience faced a 40 percent chance of reaching poverty status one or more times during their life, and low hours of paid work in midlife (less than 500 hours per year) were associated with more time spent in poverty in old age compared with women who worked at least 1,000 hours per year (Vartanian & McNamara, 2002).

Marital dissolution is another important dimension explaining elderly poverty. Elderly married couples are less likely to be poor than their single, widowed, or divorced counterparts. Many studies found that widowed and divorced older women experienced higher poverty rates than did their married peers and even unmarried older men (Choudhury & Leonesio, 1997; Crown, Mutschler, Schulz, & Loew, 1993; Weaver, 1997). In 1987, the poverty rates of older married men and women were fewer than six percent, but the poverty rates for unmarried older women were three times as high as that for married older men. Poverty rate for unmarried older women were three to four times as high as that for married older women (Burkhauser, Bulter, & Holden, 1991). Wu (2003) presented that between the years 1989 and 1993, 13 percent of older married couples spent at least one year in poverty, and 1.3 percent among them spent all five years in poverty. On the other hand, 54.2 percent of never-married older persons spent at least one year in poverty, and 11.9 percent spent all five years in poverty (Wu, 2003). Married women who experienced a dissolved marriage had a lower economic status when compared to women in a lifelong marriage (Holden & Kuo, 1996; McNamara et al., 2003). McNamara and his colleague (2003) found that among married women aged 62 or older, the poverty rate of previously divorced or widowed women was 9.6 percent and 10.8 percent, respectively, compared to 8.6 percent of women in a lifelong marriage (McNamara et al., 2003). Among non-elderly women, their low-paying jobs and occupational segregation further contributed to their poverty status (Heath & Kilker, 1992), and these economic circumstances could have a lasting impact on the economic outcomes of older women’s later life (Choudhury & Leonesio, 1997).

Living arrangement also has been a significant factor for elderly women poverty. In Korea, the poverty rate of older women living alone was 63 percent than that of married older couples (Choi & Ryu, 2003). In addition, the likelihood of older women living alone and poor was the highest and income-to-need ratio for them was the lowest (Choi & Ryu, 2003). Thus, the influence of living arrangement on older women’s poverty can be a contributing factor in the economic status of elderly women.

Finally, major determinants of economic disadvantages among older women include lower lifetime earnings, fewer years spent in the labor force, fertility experience, relatively long life expectancy, lower pension income, marital dissolution, and major health problems.

In this paper, we examine the economic status of older women using a representative sample of women aged 65 and older. The purpose of this study is to examine the income security of elderly women and the impact of different factors associated with on elderly women poverty. The rise of the aging population and the pressure of social policy changes for the elderly will directly affect the later year of the elderly live. Elderly poverty involves many complex factors including elderly individual’s attributes, life history, and various experiences across the elderly person’s life span. Additionally, elderly poverty is a controversial issue in terms of determining whether or not the elderly are a vulnerable group for obtaining social support. This study contributes to improving the understanding economic situation of older women, in particular elderly poverty, and offers considerable policy implications for preparing for elderly demographic changes in the future.

The major finding of this study is that the poverty rate for the elderly women using absolute and relative threshold in a given year is more than 40 percent. The poverty rate was particularly high as more than 70 percent of elderly were living alone. The poverty rate of older women living with adult children is the lowest, and children play an important role in providing old-age income support for their aging mothers. With respect to living arrangement, the rate of the lowest income group among WLA respondents largely increases from 68 percent to 81 percent with age. At age 75 and above, the rate of older women in the lowest income category among the WLA respondents are two times of those among the WLC respondents.

Second, the presence of income from private transfer among the WLA respondents decreases with age, even though the poverty status of the WLA respondents was severe. In contrast, the percentage having income from private transfers increases slightly if at all with age among WLS and WLC respondents. Income from family transfers is the main source of income for older women living alone and for poor elderly women, in terms of mean income share.

Third, public transfers are not significant to preclude poverty among older women. According to the result of the probity estimator, the living arrangement; the current employment status; the presence of income from assets, pension, and private transfers; net-worth levels; and educational attainment are associated with the economics security of elderly women. Living with adult children, being currently employed, and having income from investment or private transfers was significantly associated with a lower risk of poverty among older women.

Combined with the results that older women living alone are not only the most likely to have income from public transfers among the old age groups, but are also less likely to have income from private transfers among the old age groups. Also, they are less likely than others to reduce poverty once in it. Although at older ages the risk of poverty increases with age and the abrasion of asset accumulations, factors associated with older women’s poverty represent a gender-poverty linkage. The relationship between gender and economic hardship proposes certain policy implications.

First of all, income maintenance programs for poor older women need to be intensified. After controlling other variables, the presence of income from private transfers is statistically insufficient to reduce the risk of poverty. Older women have been more vulnerable to impoverishment because their assets may be eroded with the onset of chronic diseases or insufficient financial preparation for their later life. These facts imply that older women who have insecure economic conditions face intrinsic vulnerability to poverty. Thus, income support programs for poor elderly women would help to prevent older women form becoming poor.

Second, improving the life chances for younger women is needed in terms of preparing them to be economically secure in their later life. Factors such as current employment status and income from assets, pension, and/or private transfers are more likely to reduce the risk of poverty among older women. Over their life times, women are less likely to have interrupted job history and lifetime earning and their financial preparation for old ages must be secured. This is because being female is still associated with a high risk of impoverishment. In addition, the net effect of the presence of public transfers is not sufficient to preclude poverty of older women. The fact of the matter is, older women who become impoverished are certain to remain in poverty. Thus, a lifetime of work is needed in order to have secure financial independence from their marriage or caregivers within family over the course of their lifetime.