| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
a Department of Sociology and Gerontology Center, University of Kansas, Lawrence
Correspondence: David J. Ekerdt, PhD, Gerontology Center, University of Kansas, Dole Building, 1000 Sunnyside Avenue, Room 3050, Lawrence, KS 66045-7555. E-mail: dekerdt{at}ku.edu.
Decision Editor: Laurence G. Branch, PhD
 |
Abstract |
|---|
 TOP Abstract MethodsResultsDiscussionReferences |
|---|
Key Words: Retirement planning • Pension • Social Security • Health insurance
The necessary condition for anyone's retirement is income or wealth that replaces the earnings from employment. To the extent that older workers are aware of their financial prospects, retirement planning can be more complete and their decision making more confident. Ignorance about finances, however, could compromise the advantageous timing of retirement and jeopardize one's choices about activities, living arrangements, and level of consumption.
Two broad categories of knowledge equip workers for retirement financial planning. Instrumental knowledge—or financial literacy—encompasses skills for money management (e.g., how to invest, how to budget, whom to consult), familiarity with government programs (e.g., Social Security rules, Medicare coverage), and appraisals about likely conditions over the term of one's retirement (e.g., life expectancy, costs of long-term care, effects of inflation). Americans' instrumental knowledge in this regard is widely judged to need improvement, especially when moving into an era of greater personal responsibility for retirement income security. Many do not understand the basics of saving and investing, fail to take full advantage of pension opportunities, and in general are said to be undersaving for retirement (Cutler 1997
; Mitchell and Moore 1998; Moore and Mitchell 2000; U.S. Department of Labor 1998; but, see Uccello 2001). Eleven years of findings from the Retirement Confidence Survey (Employee Benefit Research Institute 2001) suggest a glass-half-full, glass-half-empty conclusion about people's financial acumen toward retirement. Adults' confidence in their retirement preparation grew through the 1990s (70% very or somewhat confident at present), and the proportion who have calculated a retirement savings goal went from 35% in 1993 to 46% in 2001. At the same time, the 2000 Retirement Confidence Survey found that accumulations were "unimpressive" as a whole and judged only 43% of workers to be doing a good or very good job of preparing for retirement (Employee Benefit Research Institute 2000).
A second category of knowledge is individuals' factual grasp of their own wealth and income potential. This is essential information for any personal calculation, estimate, or projection about personal finances—specific input for the useful exercise of generic instrumental knowledge. In this study, we will examine the sufficiency of older workers' factual knowledge about one important sector of personal finances—the retirement benefits that are earned by employment. These include the Social Security retirement benefit, savings through employer pension plans, and the eventual availability of employer-provided health insurance. Simple familiarity with one's own retirement benefits is a very basic way to evaluate the quality of American workers' preparation for retirement.
Instrumental knowledge of financial concepts is gained by experience and varies directly with education and socioeconomic status (Hershey, Walsh, Brougham, Carter, and Farrell 1998). By contrast, details about one's own employment-related retirement benefits should be readily accessible to individuals. The Social Security Administration will, upon request, furnish U.S. workers with a personal earnings and benefit estimate, and since 1999 has annually mailed such statements to workers' homes. Regulations under the federal Employee Retirement Income Security Act have required that employers periodically give eligible workers a "summary plan description" of their retirement pension and health provisions, with rights, benefits, and obligations explained in understandable language. If requested, most employers will furnish an individual benefit statement that shows the benefits earned to date and a projection of the retirement benefit. Further or updated information should be available from employee handbooks, plan administrators, or plan documents on file (Conison 1998). With workers entitled to receive information about what their plan provides and how it operates—and bureaucracies charged to disseminate it—factual knowledge about retirement benefits should not be difficult to acquire.
Hypothetically, then, individual knowledge of retirement benefits should be fairly complete. But it is not. Researchers who work with economic data have disclosed people's considerable ignorance about retirement benefits. In an analysis based on the 1983 Survey of Consumer Finances, Mitchell 1988 found that adult workers who were covered by an employer's pension had a "disturbing" lack of information about early retirement provisions. According to Curtin, Juster, and Morgan 1989, "the general view is that respondents possess little if any information about their pension rights" (p. 477). In their overview of the Health and Retirement Study (HRS), Juster and Suzman 1995 noted that "it is generally agreed among analysts that many individuals do not know a great deal about the incentive structures contained in their private pension plans" (p. S38). Smith 1995 early computation of pension wealth in the HRS had to handle "extensive missing data, especially for expected future pension incomes" (p. S178). Johnson, Sambamoorthi, and Crystal 1999 eschewed the use of "questionable self-report data" for their later study of pension wealth in the HRS. They could do so because the HRS design formally anticipated incomplete self-reporting about retirement benefits. The project has collected separate information from respondents' employers about pension and health insurance provisions, and also linked its data to Social Security earnings records (Gustman, Mitchell, and Steinmeier 1995).
Preretirees' ignorance of retirement provisions is a significant issue for three reasons. First, as noted, survey nonresponse suggests that some workers are really not adequately informed about these matters, and so lack factual knowledge to yoke with instrumental knowledge about financial management for effective retirement planning. Steps could be taken toward better disclosure of retirement benefits or toward having workers heed information already disclosed. Second, people's ignorance of retirement benefits complicates research efforts. Nonresponse to survey items creates "missing data" for researchers who want to model the financial decision set for persons who are facing retirement. "Don't know" responses that are not missing at random pose additional problems for statistical modeling, producing biased estimates of population characteristics or the size of effects (Little and Rubin 1987). Third, the conceptual significance of factual ignorance—if it is more than trivial or random—is that it calls into question assumptions that individual decision making for retirement is organized around financial factors. If some workers are not thinking closely about their finances and entitlements on the approach to retirement, then perhaps they are making decisions on some other basis.
Just how extensive a problem is workers' unfamiliarity with their retirement benefits? In this study, we examine the issue using responses to personal interview questions in the HRS, a nationally representative panel survey of persons aged 51–61 and their spouses, regardless of age (Juster and Suzman 1995). HRS survey participants had attained an age when they are more likely to be vested in pensions and also more attentive to their retirement finances than are younger adults. In this analysis, we specify percentages who cannot answer questions about pension, health insurance, and Social Security provisions, noting that nonresponse is greater for some topics rather than others. We consider whether nonresponse is randomly prevalent, or is instead related to patterns of social and occupational characteristics, as well as the temporal proximity of retirement.
The analysis is confined to responses from the baseline, 1992 wave of the HRS. As such, it describes decade-old estimates of a problem that may have been superseded by later experience. As we explain later, there are particular difficulties in using the later waves of HRS data to describe experience since 1992. At the same time, 1992 data on a representative cohort can establish a useful baseline of knowledge for later research, can inform us about the completeness of the data record in such an important panel, and can help determine whether ignorance about retirement benefits is ignorable (i.e., missing at random and not likely to bias analyses) or instead is patterned in some way.
This study focuses on people's ability to give a substantive response to questions about retirement benefits. At a very simple level, we want to know whether workers are willing to offer a response. Further evaluation of the accuracy or quality of substantive responses could be pursued in the HRS by comparing self-reports with administrative records from Social Security or employers. Gustman and Steinmeier in press conducted such an analysis and reported that, for many respondents, there were large observed discrepancies between 1992 self-reports and the detail gleaned from records. Thus, even respondents who "know" their benefits may not know them reliably, and so the present focus on the don't know response, if anything, probably underestimates the level of ignorance about retirement benefits.
|
Methods |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
When households were used as the analytic unit of the HRS—with one "financially knowledgeable" respondent speaking for the household—the same filtering rules as described reduced the full number of household respondents from 7,706 to 3,406. In this set, the financial respondents were 43.2% female, 82.3% White, had a mean age of 55.6 years, and had 87.2% of respondents working 35 or more hours per week.
Question Series About Retirement Benefits
In face-to-face interviews, HRS participants who were workers were steered through questions about employers' pensions, and the financial respondents were also asked about Social Security retirement benefits and employer-sponsored health insurance. During the interviews, HRS respondents were allowed, but not actively encouraged, to consult documents they might have in the house. Also, the HRS was fielded during the spring tax season, when people might have a better idea of their financial situation (HRS staff, personal communication, June 28, 2001).
Employer Pensions.
Workers were directed into a series of questions on pension plans if they participated in a plan sponsored by their employer or union. Self-employed workers, too, answered these series if they were included in "any pension plans or tax-deferred savings plans through your work." Respondents were further directed into separate series depending on their participation in defined benefit ("formula-type") plans or defined contribution ("account-type") plans. The latter type was defined by HRS as encompassing 401(k), 403(b), ESOP, SRA, thrift/savings, or stock/profit-sharing arrangements. Table 1 lists most of the questions in each series, excluding some items that followed up specific answers. In both series, some questions concerned personal intentions for pension receipt and other questions concerned general plan features. For researchers interested in retirement planning, the responses of greatest interest would be those about eligibility ages and about pension values. To reduce the amount of missing data in the reporting of expected benefit amounts or account balances, respondents were allowed multiple ways to answer: as amounts per month or per year (exactly or selected from ranges shown on a card), as a lump sum, or as percentages of final pay.
|
Continuation of Health Insurance.
Financial respondents were also asked questions about health insurance. Those in households with any type of coverage were asked whether the coverage—be it from one spouse or the other—was available to people who retire. Those with continuation coverage were also asked about the employer or union's share of the costs, and whether spouses could be likewise covered.
Note that questions about employer pensions concern benefits on the current, main job, not pension entitlements from former jobs. We also confine the analysis to details of the "most important plan."
Other Variables
To determine whether don't know responses were patterned, selected benefit items were compared on other variables. Temporal proximity to retirement was one such variable because ignorance of retirement benefits may be confined to workers who are far from the event, and so the problem ultimately resolves itself. Each of the question series included items about an expected age for initiating benefits, from which temporal proximity was calculated.
Selected don't know responses were also compared on a set of social and occupational characteristics of the opportunity structure for retirement planning. They indicate the degree to which a worker has a life that is socially "attended" in a way that makes one's retirement intentions a public matter—on the job, at home, or among friends (Ekerdt, Hackney, Kosloski, and DeViney 2001). An important part of the attended life is access to bureaucratic employment, and to the information, job security, and material resources that such employment provides for retirement planning. We would expect that workers with less attended lives would know less about retirement benefits, and so we expect that don't know responses are more likely among women, non-Whites, persons with less education, part-time workers (<35 hours per week), and workers who do not belong to unions. Away from the job, having a marital partner and having health limitations should raise the everyday social topicality of retirement, and so don't know responses should be less likely under these circumstances.
|
Results |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
Employer Pensions
Of the 4,980 workers in the study population, 2,888 (58%) said that, on their main job, they were included in a pension, retirement, or tax-deferred plan specifically sponsored by an employer or union or "through your work." A small number of workers (30) did not know whether they participated in any plan. The 2,888 claiming to have a pension plan were next asked to identify the "most important plan" either as a formula type ("benefits usually based on a formula involving, age, years of service, and salary") or an account type ("money is accumulated in an account for you"). We will refer to these, respectively, as defined benefit (DB) and defined contribution (DC) plans. Altogether, 1,722 (61.4%) identified a DB plan as most important, and 977 (33.8%) identified a DC plan as most important. Persons who said that their plan had features of both types (3.3%) or did not know the type (1.5%) were sent into the question series on DB pensions (resulting in 1,910 cases in all). Gustman and Steinmeier in press warn that even these basic responses about plan type may not be reliable. Possible misunderstanding about plan type may partially account for workers' further ignorance of specific plan features.
Sections A and B of Table 1 report acquaintance with details of the most important plan. For both types of plans, workers were nearly completely cognizant of the number of years they have been in the plan, which is a fact of one's employment history. The great majority also cited an expected age for start of benefits. Only 8.3% of the DB participants and 11.3% of the DC participants had not decided when they will elect to begin benefits, proportions that conform to other HRS information about uncertain plans for the timing of retirement (Ekerdt et al. 2001).
Looking within the set of DB questions (Section A), a key input for retirement decision making is the age that one is eligible for benefits. Respondents were asked the earliest age at which they would be eligible for full benefits, which 7.0% did not know, and the earliest age for any benefits, which 8.5% did not know. Slightly higher proportions did not know about such plan conditions as their contribution amount (if any), integration with Social Security, or lump sum settlements.
The highest levels of ignorance were coded for expected levels of pension income. Fully 34.6% could not state their own expected benefit as an amount per month or year, or as a lump sum, or as a percent of final pay. Of those who could answer the question about the normal (full-benefit) retirement age, 36.2% did not know what the benefit amount would be. Of those with an early, reduced-benefit age, 37.7% could not characterize the reduction or say whether there even was a reduction.
Ignorance of one plan provision was related to ignorance of others. For example, those who did not know the plan's normal retirement age had 3.1 higher odds of not knowing about contributions, and 4.1 higher odds of not knowing one's expected benefit amount.
Altogether, awareness of DB plan features is impressive in many respects, with about 90% or more knowing operational details. However, the 7%–8% of participants who cannot identify the early or normal retirement ages are lacking key information. The plan features that have high cognizance here are the ones that are administratively disclosed to workers—the kind of public knowledge that is likely to be topical conversation in the workplace. Ignorance ran higher—around 35%—about potential benefit amounts or levels. These are projections that are specific to the individual and that are generated by a formula applied to the worker's record. Such information takes more personal initiative to acquire and remember.
Similar patterns of substantive versus don't know responses were found among DC plan participants (Section B). DC plans require more individual agency than DB plans, and the cognizance of plan operations was equally high for both pension types. Nearly everyone knew about the opportunity to choose investments, and more than 90% of participants offered a substantive response about plan type or could speak about lump sum or installment distributions. Awareness of contribution amounts varied: about 6% did not know their personal contribution, but 20% did not know the employer's or business' contribution. As for personal reporting about the size of one's account, ignorance ran higher than among all other questions: 20.4% could not say how much money was presently in their accounts. This don't know percentage about pension wealth was lower than among DB participants; but, in fairness, DB participants were asked for a projection of their retirement wealth, and DC participants were asked for a current account balance. As with DB plans, don't know responses were not random in the DC series. For example, those who did not know their employer's contribution level had odds 4.1 times higher of not knowing their account balance.
Sections A and B of Table 1 report workers' responses about their "most important" plan. Perhaps workers with multiple plans might be more knowledgeable about their second or third plan, but this was not likely (data not shown). For example, persons who did not know the pension value of their first plan (expected benefit of a DB plan or account balance of a DC plan) had 4.0 higher odds of not knowing the pension value of their second DB or DC plan.
The question series about employer pensions revealed that workers readily offer responses about the shared, public details of the plan, but knowledge that they could use to estimate or foresee pension wealth was lacking among one third of DB plan participants and among one fifth of DC plan participants. Even at this level of item nonresponse, HRS respondents were more informed than the adult sample of the 1983 Survey of Consumer Finances, in which Curtin and colleagues 1989 found that the question on expected pension amount had a missing data rate of 61%. At the same time, individuals' ignorance of pension values in the HRS was higher than household ignorance of other asset values, where don't know responses (by the financial respondent) were in the single digits for the value of the house, stocks, and checking accounts (Smith 1995).
Retiree Health Insurance
The set of questions about retiree health insurance was answered by all individuals without partners and by the "financially knowledgeable" person in a two-partner household. Of these 3,406 nonretired respondents, 2,725 reported health insurance coverage from their own (or spouse's) employer or union. Those with coverage (Table 1 , Section C) were asked whether their household's "first" health plan would be available to retirees, a provision that 14.0% did not know. (Among the 280 people with a second health plan, 20.4% did not know about that plan's coverage for retirees.) Among respondents who said yes—that coverage was available to retirees—11.6% did not know how costs were met, and 10.1% did not know about spouse coverage. Thus, about 1 of every 7 or 8 (preretired workers in) households with employee health insurance was proceeding without clarity about continuation coverage. Of those who said they have continuation coverage, about 1 in 10 did not know important details.
Social Security
Again, only one person per household was asked the brief question series about anticipated Social Security receipt. Of these 3,406 respondents—all working, aged 51–61, and claiming not to be retired—3,342 were not current recipients of Social Security in 1991, but expected to be at some time in the future. These individuals form the respondent set for questions about Social Security. In Table 1 , Section D, we have shown responses only to questions about the respondent's own Social Security plans, not those of any partner.
As for age of expected benefit receipt, 95.5% gave a substantive response. This is a higher proportion than offered an expected age for beginning the DB (90.8%) or DC pension (86.7%), but then again the set of possible ages to choose from is probably narrower for Social Security, not beginning until age 62. Regarding expected amount, this knowledge was lower than for employer's pensions. Half of potential beneficiaries (51.9%) could not offer an expected amount, even if using a range card and even though identified to be the household's financial specialist.
Among Those Closer to Retirement
Workers' familiarity with retirement benefits is likely to be greater to the extent that they see themselves closer to the event and for two reasons. First, individuals themselves may seek out, acquire, and retain more information as they approach their time for retirement. Second, the social environment may offer or ask more information when workers are known to be near-retirees. Retirements are consequential to spouses, family members, friends, coworkers, supervisors, customers, and clients—all of whom can raise the topic of an individual's plans and force attention to details.
Don't know responses for selected benefit questions are shown in Columns A–D of Table 2 , grouped by years of self-reported temporal proximity to benefit receipt or retirement. Proximity was determined as follows. Three of the four HRS question series about retirement benefits—DB plan, DC plan, Social Security—included a question about the expected age for benefit receipt, which was used to calculate years of proximity (see footnote to Table 2 ). For health insurance, proximity was calculated from a general question about expected age for retirement. Proximity was grouped into 3-year categories, with a separate category for workers who did not know when they would begin benefits or retire.
|
The proximity gradient seen for three items in Table 2 remains even when covarying for factors shown by prior research to be related to retirement plans (Ekerdt et al. 2001; Kosloski, Ekerdt, and DeViney 2001). Table 3 reports findings on the same four selected benefit questions, where logistic regression models were used to predict the don't know response to each question. In this multivariate setting, workers further in time from retirement generally had higher odds ratios predicting the don't know response relative to the reference group of workers 0–3 years proximate. The pattern held for don't know responses about defined benefit pension amounts, health insurance continuation, and Social Security payments (Models A, C, and D).
|
From the pattern in Table 2 and Table 3 , one would infer that unfamiliarity with retirement benefits should also diminish over time. Time, however, does not necessarily bring individuals closer to retirement, especially those who switch to later exit intentions. But time does bring people more experience as pre-retirees and perhaps more knowledge. The longitudinal trend in don't know responses might be traced through subsequent biennial waves of the HRS panel were it not for difficult issues in explaining and interpreting change. Indeed, there could be as many as four bases for changing rates of nonsubstantive response about retirement benefits. An aging effect, as noted, should induce greater familiarity. A period effect could likewise induce greater familiarity, as when increasing financial literacy in the public as a whole makes workers more attentive to retirement provisions. A selection effect could leave the cohort of continuing workers biased in some unknown way (i.e., selective retirements could remove the more informed respondents). Finally, respondents could become more informed by the sheer fact of HRS-induced sensitivity to details that were asked about in repeated surveys. An examination of don't know responses in the 1996 wave of the HRS (the most recent public data available) shows that unfamiliarity did diminish after 4 years of follow-up. An analysis to determine whether new levels were from age, period, selection, or testing effects is beyond the scope of this study.
|
Discussion |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
These don't know responses occurred even after the HRS went to impressive lengths to minimize item nonresponse. Nor were these answers missing at random. Not knowing one detail was associated with not knowing others. An analysis of selected items showed that workers who, by their own reckoning, were more temporally remote from retirement, were less likely to know details. Even holding proximity constant, don't know responses were more likely among workers who were subject to less social and bureaucratic supervision of the retirement process.
Ignorant of their benefits, some numbers of pre-retirees have an impaired capacity for financial decision making around retirement. But that is only part of the problem. Even those HRS respondents who gave affirmative answers may be misinformed about their benefits (Gustman and Steinmeier in press). Even accurate information cannot be put to good use without financial literacy, and Americans' financial literacy for retirement planning is worrisome to experts (Cutler 1997; U.S. Department of Labor, 1998).
The prevalence of don't know responses is far more than a problem of missing data, but as missing data, don't know responses are problem enough. For example, should researchers want to use a "pension wealth" variable to model individual decision making or to estimate the population's savings for retirement, its computation requires a complex formula plugging in detailed information from the kinds of questions shown in Table 1 . If answers are not available or if self-reports are not trusted, then researchers have some recourse—they can replace, impute, or assume the missing values. For just this purpose, the HRS anticipated incomplete or inaccurate self-reports. Project staff have solicited administrative records from respondents' employers and from the Social Security Administration, and these supplemental data have allowed computations of pension wealth to proceed (e.g., Gustman, Mitchell, Samwick, and Steinmeier 1999; Johnson et al. 1999; McGarry and Davenport 1998; Smith 1995). However, due to incomplete coverage, records can only be a partial solution to missing self-reports. For the 1992 wave, Social Security earnings histories will be available for perhaps 80% of respondents, employer-provided pension plan descriptions were matched for 67% of cases, and health insurance plan descriptions were matched for 72% of cases (Gustman et al. 1995).
Yet even if analysts can simulate omniscient financial actors—with all disclosure of attendant assumptions and biases—the fact remains that missing data signifies real missing knowledge in the population, as well as the planning behavior that follows from it.
These findings argue for a theoretical reconsideration of the role of financial knowledge in retirement behavior. Nonnegligible proportions of workers in their 50s are walking around without the basic input for a simple retirement financial calculator, such as the American Savings Education Council's "Ballpark Estimate" (www.asec.org) or Money magazine's "Retirement Savings Calculator" (www.moneymag.com). The personal knowledge and acumen that are idealized in behavioral models of retirement (Lumsdaine and Mitchell 1999) or advertisements for retirement financial planning (Ekerdt and Clark 2001) seem beyond the experience of many workers. The challenge, then, is to specify how such incentives as pensions and health insurance figure into decision making. Some workers may weigh the financial advantages of work versus retirement and time their exits to maximize present or future income. Some, however, may merely imitate the informed decisions of savvy coworkers (Axtell and Epstein 1999) or let their spouses figure it out. For others, it may be the simple availability of a pension, more so than a threshold amount, that creates the incentive to retire. People live and work within structures that will provide them with retirement income (employers' pensions, Social Security) and that bear them along with suggestions (norms) for the timing of the event (Ekerdt 1998). Many retirements may be less a matter of rational agency than a compliance with these structural arrangements, accepting the circumstances that one has been dealt, and trusting that the eventual income will suffice. Replacement income may be the necessary condition of retirement, but perhaps not a necessary feature of decision making.
It could be argued on two grounds that ignorance of retirement benefits is a minor problem. First, perhaps some workers feel so well fixed financially for retirement that details need not matter. Although it is possible that some can "afford" to be ignorant, our findings showed that don't know responses were actually less likely among those with more education and presumably more affluent. McGarry and Davenport 1998 also report that HRS workers with more generous pensions had more complete responses on pension questions. Second, it is reasonable to suppose that people with information deficits (at time of HRS interview) could brush up on their benefits when the need arose. Details about retirement benefits could be looked up in time for immediate decision making. Yes, they could, for such information, as noted previously, is generally accessible. Yet, absent walking-around knowledge of one's retirement benefits, people probably do not routinely process such questions as "Am I saving enough for retirement?" or "What is my risk tolerance?" They probably do not factor in serendipitous information about insurance, investments, and taxation, nor do they attend to the entreaties lodged in retirement financial planning advertisements. When people do not know their financial condition, they forego everyday opportunities to measure their preparation for the future.
As a practical matter, better knowledge of retirement benefits should flow from stepped up disclosure practices. For example, Social Security's new, annual benefit statement mailouts should reduce the high ignorance of benefit amounts that was recorded among HRS respondents in 1992 (Efforts to Inform the Public, 2000). Beyond the provision of information, others urge public exhortations to compel attention to retirement preparation (U.S. Department of Labor, 1998), although the effectiveness of such campaigns can be questioned (Anderson, Li, Bechhofer, McCrone, and Stewart 2000). Personal and computer-aided counseling are perhaps the best way to make workers attentive to their financial prospects. In addition, adults' misinformation about retirement benefits can only be good news to the willing corps of financial planners.
Yet, even after a decade of quickening popular attention to retirement (Ekerdt and Clark 2001), only half of American workers say that they have calculated a retirement savings goal (Employee Benefit Research Institute 2001). People's differential inattentiveness to retirement benefits and finances begs for research attention. Is it some interaction of situation and personality? Etzioni 1993 argues that, indeed, many choice behaviors are determined by "normative affective" factors rather than a careful analysis of information. In the normative affective model, decisions are made by individuals largely on the basis of a combination of normative frameworks and emotional values. Today, there exists an attitude of givenness toward the eventuality of retirement, and the closer one is in time to the event, the more topical engagement one tends to have with retirement (Ekerdt, Kosloski, and DeViney 2000). Retirement is not only a "natural" feature of the life course, it is often positively valued, especially when daily routines of work are unsatisfying. If, indeed, people are prone to make retirement decisions on the basis of normative affective criteria rather than thorough analysis and calculation, we must carefully evaluate the policy implications of continuing to view people as rational actors. As segments of the policy community call for further moves away from structural arrangements for retirement income, such as defined-benefit pensions and the guaranteed Social Security benefit, and toward more voluntary saving with more personal initiative—favoring wider options, numerous decisions, and the need to monitor more information—it is most pertinent to ask how well informed, prepared, and inclined people are to undertake these tasks.
The Forum
Book Reviews
Practice Concepts
|
Acknowledgments |
|---|
Received for publication August 31, 2001. Accepted for publication December 28, 2001.
|
References |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  V. Phua and J. W. McNally Men Planning for Retirement: Changing Meanings of Preretirement Planning Journal of Applied Gerontology, November 1, 2008; 27(5): 588 - 608. [Abstract] [PDF]
|
|
|
|
 |
|
 D. A. Hershey, K. Henkens, and H. P. Van Dalen Mapping the Minds of Retirement Planners: A Cross-Cultural Perspective Journal of Cross-Cultural Psychology, May 1, 2007; 38(3): 361 - 382. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
