CULTURAL DIMENSION OF THE DISCREPANCY OF PLANNED AND ACTUAL RETIREMENT AGE IN THE U.S.: HISPANICS VERSUS NON-HISPANICS

Abstract The sustainability problems to the SS has led to a glowing debate about what the full retirement age should be and if working longer is a plausible option for everyone or just for those who have some control over their retirement decisions (e.g., Munnell & Sass, 2008; McNamara & Williamson, 2013; Munnell et al, 2016). All ethno-racial groups have increased their average retirement age over the last years. However, Hispanics’ retirement age is still lower even if they stated they plan to continue to work at retirement (EBRI 2008; Diaz-Valdes, 2018). Most studies about retirement timing have focused on middle-class Whites, and the prediction of planned or actual retirement separately. One of the lesser studied complexities of the retirement conundrum concerns ethno-racial differences and cultural-related predictors of retirement timing (Lytle et al, 2015). This study seeks to extend the understanding of differences between Hispanics and non-Hispanics regarding the timing of retirement relative to when they thought they would retire by including a broad array of cultural and family related predictors. Multinomial regression models were used. The results indicate significant differences between Hispanics and non-Hispanic Whites. Taking care of grandchildren was a significant predictor among Hispanics but not among non-Hispanic Whites. For Hispanics taking care of grandchildren, for over 20 hrs., was associated with a decreased probability of stating they will never retire. The increase of one dependent was associated with an increased on the probability of retiring earlier than planned. The effect of one additional dependent was larger for non-Hispanics.

in countries with mandatory retirement systems, since here employer and employee need to negotiate a new contract after normal retirement age. I study employers' motives to and experiences with employing older workers after normal retirement age using data from a 2017 survey among 1,312 Dutch employers. The Netherlands has mandatory retirement regulations but is also seeing an increase in employment rates after normal retirement age. Results show that 54% of employers have, in recent years, employed one or more older workers beyond their normal retirement age. This is especially common in education. 70% of employers are very positive about their previous experiences with employing older workers after normal retirement age, mostly because they had rehired older workers with unique knowledge and experience. However, employers also hardly ever took the initiative for such employment arrangements, instead leaving it to the older workers to show the desire to continue working. Although employers are largely positive, they see it as a limited phenomenon, and do not consider it a solution to labor shortages.

CONSEQUENCES OF AGE DISCRIMINATION IN THE WORKPLACE
Dannii Yeung 1 , 1. City University of Hong Kong, Hong Kong, Hong Kong Despite a growing number of middle-aged employees in the workforce, Hong Kong still lags behind other developed countries in implementing age-friendly policies to protect older workers from being discriminated in recruitment, performance appraisal, and other personnel decisions. This paper therefore aims to investigate the effects of age discrimination on work-related outcomes and well-being. A total of 333 Hong Kong Chinese employees aged 40 years and above (Mage = 46.6 years, SD = 6.21, Range = 40-68; 60% female) completed an online survey on work experiences. Self-perceived age discrimination in the workplace (Furunes & Mykletun, 2010), perception of occupational future time (Ho & Yeung, 2017;Zacher & Frese, 2009), work engagement (Schaufeli et al., 2017), work stress (Cohen et al., 1983, and psychological distress (Shek, 1989) were measured. Almost 53% of the participants reported that older workers were less likely to get the same training opportunity for new technology as their younger peers, and 46.8% perceived that older workers did not get the same promotion opportunity as their younger peers. Regression analyses further revealed that employees' perceived age discrimination in the workplace indirectly influenced their levels of work engagement (B = -.038, SE = .017), work stress (B = .016, SE = .005), and psychological distress (B = .008, SE = .005) through perception of occupational future time. The findings of this study unveil the severity of workplace age discrimination in Hong Kong and the negative impacts of age discrimination on workrelated outcomes. Recommendations for organizational practices will be discussed. In the context of an aging society, where the proportion of older adults is rapidly increasing, ensuring healthier longer lives is key for individuals, families, policy makers and the population as a whole. In this context the productive aging framework has gained increased importance. There is evidence showing that engagement is related to late-life well-being and health (i.e., Hinterlong, 2006;Everard et al, 2000;Rozario et al, 2004;Matz-Costa et al, 2012). However, the productive aging framework lacks cultural sensitivity and evidence about the association between the effect of retirement on health and well-being in late-life is mixed. The current study seeks to contribute to this gap by exploring the consequences of the discrepancies between planned and actual retirement age on subjective health and well-being, comparing Hispanics and non-Hispanic Whites. A series of regression models were conducted to explore the effect of the discrepancy between planned and actual retirement age on retirement satisfaction, self-rated health and mental health (CESD). Results indicates that native born Hispanics presented more differences when compared to foreign born Hispanic than non-Hispanic Whites, which could indicate the effect of acculturation and its fading effect on cultural attitudes, such as familismo. Hispanic tend to have higher retirement satisfaction than non-Hispanics which is aligned with the happiness paradox found by Calvo and collagues (2017). Additionally, SES has a significant effect on health for non-Hispanic Whites but not among Hispanics. Finally, retirement timing predicted mental health among foreign born Hispanic but among native born Hispanics and non-Hispanic Whites. The sustainability problems to the SS has led to a glowing debate about what the full retirement age should be and if working longer is a plausible option for everyone or just for those who have some control over their retirement decisions (e.g., Munnell & Sass, 2008;McNamara & Williamson, 2013;Munnell et al, 2016). All ethno-racial groups have increased their average retirement age over the last years. However, Hispanics' retirement age is still lower even if they stated they plan to continue to work at retirement (EBRI 2008;Diaz-Valdes, 2018). Most studies about retirement timing have focused on middle-class Whites, and the prediction of planned or actual retirement separately. One of the lesser studied complexities of the retirement conundrum concerns ethno-racial differences and cultural-related predictors of retirement timing (Lytle et al, 2015). This study seeks to extend the understanding of differences between Hispanics and non-Hispanics regarding the timing of retirement relative to when they thought they would retire by including a broad array of cultural and family related predictors. Multinomial regression models were used. The results indicate significant differences between Hispanics and non-Hispanic Whites. Taking care of grandchildren was a significant predictor among Hispanics but not among non-Hispanic Whites. For Hispanics taking care of grandchildren, for over 20 hrs., was associated with a decreased probability of stating they will never retire. The increase of one dependent was associated with an increased on the probability of retiring earlier than planned. The effect of one additional dependent was larger for non-Hispanics.

EVALUATING THE FEASIBILITY AND EFFICACY OF A PILOT INTERVENTION TARGETING UNEMPLOYED OLDER WORKERS Jie Yang 1 , 1. East Carolina University School of Social Work, Greenville, North Carolina, United States
The great recession of 2007-2009 has led to historically high unemployment rates, and almost half of the unemployed population was long-term unemployed in 2010. Both discriminations against the long-term unemployed and ageism hinders older unemployed workers reentering the workforce, and they risk severe mental health deterioration. However, traditional job training programs are shown to be ineffective in protecting their participants' mental health. Hence, adopting Jahoda's Latent Deprivation Theory, the author worked with a non-profit in the Greater Boston area on developing an innovative intervention targeting older unemployed older workers' mental health and income generation outcomes. The pilot intervention (lasting three months) created a quasi-office environment where participants can have a time structure, engage with peers, work on projects, and receive training. The current study evaluates the effect of this pilot program on multiple mental health outcomes. Pre-and-post comparisons were conducted using quantitative analysis (N=13). Participants' depressive symptoms dropped significantly, and they reported having a better time structure of the day as well as higher scores in extraversion in terms of personality change. Qualitative analyses were conducted to analyze the pre-and-post interviews. Participants reported overwhelming approval of the intervention in helping them better cope with unemployment and being productive in seeking jobs or generating incomes. One example of quotes is "…compared to three months ago, I'm in a very different place. Before the Collaboratory, I had become extremely depressed and withdrawn…The Collaboratory has helped make a massive difference in my sense of well being right now."

AFFECT MODERATES THE ASSOCIATION BETWEEN SOCIAL SUPPORT AND RETIREMENT SATISFACTION OVER TIME Kaleena Odd 1 , 1. University of Nebraska Omaha, Omaha, Nebraska, United States
Retirement is becoming more important for today's older adults because they are living longer than before. Recently, research has started to explore how different individual resources (e.g., health or finances) and social resources (e.g., social support or social network size) influence retirement outcomes such as retirement satisfaction. Moreover, the current study sought to examine the influence of time, satisfaction with social support, and affect (i.e., positive or negative) as predictors of retirement satisfaction. Data was obtained from a longitudinal study that explored how older adults in Montreal, Canada adjusted to life in retirement over the course of three years. Hypotheses were tested using a structural equation model that investigated retirement satisfaction as predicted by time, satisfaction with social support, positive affect, and negative affect. Gender differences were also explored. Overall, there was no change over time among the variables. Satisfaction with social support, positive affect, and negative affect were all associated with retirement satisfaction in the expected directions. Positive affect moderated the association between satisfaction with social support and retirement satisfaction, such that the association was stronger for those low in positive affect. Also, negative affect moderated the association between satisfaction with social support and retirement satisfaction as a function of gender. This study extended the literature by exploring how multiple predictors interacted to influence retirement satisfaction over time. Future research should examine how individual and social resources can interact with each other to better understand retirement satisfaction.

PRECARIOUS PATHWAYS INTO RETIREMENT AND NEW RISKS FOR GENDERED ECONOMIC EXCLUSION IN SWEDEN, 1990-2015
Andreas Motel-Klingebiel, 1 and Susanne Kelfve 1 , 1. Linkoping University, Linkoping University, Sweden, Sweden The ability and disposition of ageing people to maintain their labour market activity and/or to retire from work structurally depend on pension systems, activation policies, ageism, changing for labor demand and economic shifts. Structural conditions are changing, but social change does not mature homogeneously and neither do the institutional shifts induced by it. Gains in opportunities and resources do not benefit all people, groups and even societies in the same way. Changes increase insecurities and life course inhomogeneity, create unequally distributed challenges and show asynchrony in shifts and outcomes. They generate new precarity in ageing and socially structured risks for exclusion in work and retirement and refer to existing later life inequalities by cohort, gender, region, education, class and ethnicity. From this perspective of ageing and social change, the paper deals with shifts in late work and retirement patterns and laterlife outcomes under changing institutional conditions, focusing on gendered risks for economic exclusion and later life precarity in Sweden. Swedish registry data comprising individual work and health histories as well as employer, regional and neighborhood information on the total population 50+ ever living in Sweden 1990-2015 is used in a cohort sequential perspective. Analyses focus on gender inequalities and concentrate on occupational activities, retirement transitions and pension revenues under changing social conditions. Models find increasingly heterogeneous preretirement and transition patterns, new gender gaps and increasing risks of economic exclusion in retirement with disadvantaged groups as forerunners in overall relative declines in later-life economic positions.