Retirement Plans: Planners and Nonplanners

Abstract

With the aging of millions of baby boomers, retirement planning is rapidly increasing in importance. This study explores differences in characteristics of individuals who have well-articulated retirement plans and those who do not. Women and those with low financial net wealth are less heavily represented among planners. More heavily represented among the planners are the self-employed, those with high financial net wealth, and those who have spoken to a professional financial planner. The expected primary source of retirement income is also associated with planning status. Those expecting to rely primarily on Social Security for their retirement income are less heavily represented among planners, while those expecting to rely mostly on defined-contribution plans or on savings or income from personal investments are more heavily represented. Since procrastination, inertia, and low expectations of success often impede retirement planning, three recommendations are provided: (a) develop user-friendly software to entice individuals to explore retirement planning in a fun way; (b) improve financial literacy education to develop understanding and motivation; and (c) institute programs that use inertia to encourage greater saving.

The 75 million U.S. baby boomers, born between 1946 and 1964 (Goldstein & Damon, 1993), are rapidly reaching traditional retirement age. The oldest of the boomers will be eligible for full Social Security benefits at age 66 in 2012 (Social Security Administration, 2010), and many have already taken the reduced Social Security benefits that become available at age 62. A 2008 survey found that 45% of individuals who were 61 years old planned to take Social Security benefits at age 62 (Fidelity Investments, 2008); their reasons included financial needs and health concerns. In 2009, the number of new Social Security beneficiaries, aged 62 and older, reached a record high with 1.3 million men and 1.9 million women receiving benefits (Johnson & Mommaerts, 2010). Alan J. Auerbach, professor of economics and law at the University of California at Berkeley, observed that many elders who want to keep working are unable to find work and, therefore, are forced to take early retirement (Associated Press, 2009).

Many boomers do not have well-articulated retirement plans. In a recent study of Pennsylvania workers born in 1964 or before, Brucker, Leppel, and Cochrane (2008) reported that only 44% of respondents indicated that they had a financial plan with specific goals for retirement. Even among those within five years of expected retirement, only 50% had a plan. Based on a study of individuals over age 50, Lusardi and Mitchell (2006) found that 31% of respondents had attempted to determine what they needed to save for retirement. Of those who tried to do so, only 67% actually created some sort of a plan. Of those who developed a plan, 88% essentially stuck to the plan. So for their sample as a whole, Lusardi and Mitchell found a 19% overall rate of successful planning. As they concluded, “planning for retirement is difficult, few do it, and fewer still think they get it right” (p. 12).

Making a bad situation worse, stock market declines, decreasing home prices and increasing retirement health costs have made many previously barely adequate retirement plans no longer consistent with the dream of maintaining the same standard of living after retirement as before. The National Retirement Risk Index (NRRI), which is calculated by the Center for Retirement Research at Boston College, indicates that 51% of working-age American households are at risk of being unable to maintain their standard of living in retirement (Munnell, Webb, & Golub-Sass, 2009).

For some older boomers, it may be too late to plan to significantly improve their expected standard of retirement living through increasing their retirement savings. However, for many other younger boomers, there is the possibility of doing so. Some might be able to increase their standard of retirement living by working longer than was originally planned. For others, working longer may not be a viable option given health issues. Furthermore, with a weak economy and jobs scarce, some workers who might want to work longer than originally planned may not have that option. They may, therefore, apply for the reduced, early Social Security benefits. Some individuals may have no choice but to accept the lower levels of replacement retirement income provided by their 401(k) plans that have been depleted by stock market declines.

Given the economic climate and the number of baby boomers poised on the edge of retirement, understanding factors related to retirement planning is of particular importance. With 15.3% of its population over 65 years of age, Pennsylvania has the third oldest population in the United States (U.S. Census Bureau, 2009). Thus, the state has a wealth of information to offer on the attitudes and behaviors of older individuals. The current study uses data collected by the Widener Elder Pennsylvanian Survey (WEPS) to explore the differences between planners and nonplanners. This study finds that the expected primary source of retirement income is related to the likelihood that an individual has a well-articulated retirement plan. Gender and net worth are also associated with whether an individual has a clear retirement plan.

THEORETICAL BACKGROUND

According to traditional life cycle theory, individuals are rational planners of their consumption needs over their lifetimes. During their youth, they consume more than they earn, financing the difference by borrowing. The middle-aged years are an accumulation phase during which individuals stockpile for retirement. During old age, they decumulate savings. To save appropriately, individuals must solve present value problems based on their approximation of unknown facts about the future such as earnings, asset returns, tax rates, health status, and life expectancy (Mitchell & Utkus, 2004).

An alternative theory states that when individuals operate in complex and uncertain situations, their decisions are subject to bounded rationality rather than classical omniscient rationality (Simon, 1979). Decisions are further limited by bounded willpower; saving for retirement is difficult both cognitively and in terms of self-control (Mullainathan & Thaler, 2000). Empirical evidence indicates that most people do not plan well for retirement. Only 44% of U.S. workers (or their spouses) have calculated how much they need to save for a comfortable retirement and 25% have no retirement savings (Helman, Copeland, & VanDerhei, 2009).

Selnow (2004) provided explanations for why people tend not to plan well for retirement. First, the payoff for retirement saving is uncertain and distant. Next, the cost of retirement saving is foregone current consumption; thus, failure to save yields instant gain. Furthermore, there is no immediate reward for doing the right thing, and no immediate penalty for not doing it. Last, Selnow explained, there are no specific deadlines before which one is fine and after which one is in trouble.

Also preventing individuals from taking appropriate actions is the power of inertia. To help increase worker savings and retirement preparedness, some firms have turned to automatic enrollment of new employees in company 401(k) plans. The research of Choi, Laibson, Madrian and Metrick (2001a, 2001b) suggests that these automatic enrollment policies can influence employees’ retirement savings. While employees can opt out of the 401(k) plan, relatively few do so. Nessmith, Utkus, and Young (2007) reported that under automatic enrollment, plan participation was 86% compared to only 45% when enrollment was voluntary. The difference is particularly dramatic for low-income workers. For employees earning less than $30,000, those with automatic enrollment had a participation rate of 77% compared to only 25% under voluntary enrollment. Unfortunately, most employees accept the low default savings rate. Thus, the effects on savings of automatic enrollment policies may operate in opposite directions for different individuals. Savings are increased for those who would not have participated in the 401(k) plan were it not for the policy. However, the low default values may lead some participants to opt for lower savings rates than they might have chosen had they made their own selections. The power of inertia is further evident in that years later many plan participants continue to contribute at the default rate.

Thaler and Benartzi (2004) demonstrated that inertia can be used to induce greater saving. Under the Save More Tomorrow^TM plan, workers agreed in advance to allocating part of their future salary increases to retirement savings. Then, action was not required to increase saving; action was needed to not increase it. After the program had been in place for four annual raises, the findings showed that a high percentage of those offered the plan joined, and the vast majority remained enrolled through all four raises.

Given the problems of inertia and other issues raised by Selnow (2004), it is almost surprising that any retirement saving does occur. So why is it that some people plan and save and some do not? The current article hypothesizes that planners and non-planners differ in logical and explainable ways. In particular, by planning, planners demonstrate the importance they place on the future. Hence, the rate at which they discount the future may be less than for nonplanners.

In his theory of planned behavior, Ajzen (1991) stated that while stronger intention or motivation increases the likelihood of a behavior, the individual must also perceive the behavior as under his/her control. The individual must see him/herself as having available the “requisite opportunities and resources (e.g., time, money, skills, cooperation of others)…” (p. 182). In the context of retirement planning, opportunities and resources are needed on two levels. First, the individual must be capable of developing the plan. If a person has little understanding of personal finance, no knowledge of someone who could assist him/her, and few financial resources to hire a professional with the necessary skills, the development of a plan is highly unlikely. Second, the individual must believe that he/she will be able to follow through on the plan. A person is unlikely to expend the effort and money required for retirement planning if he/she believes that the plan has minimal chance of success. This second part is consistent with Ajzen's statement that perceived behavioral control influences both intention/motivation and behavior.

A number of variables were hypothesized as being associated with whether an individual has a well-articulated retirement plan, by affecting either motivation or perceived behavioral control. Age is one such variable. Individuals who see retirement as rapidly approaching may see a pressing need to plan for that retirement.

Greater net worth may reflect characteristics associated with low discount rates such as patience and a willingness to delay gratification (Lusardi & Mitchell, 2007), which may increase an individual's disposition to develop a plan as well as the likelihood of having a successful plan. The future-looking attitude of those with low discount rates may provide the motivating force needed to overcome inertia and create an effective financial plan. Those with greater net worth also have more wealth with which to work and increased incentives to optimize tax advantages. On the other hand, those with less wealth may feel greater need to plan to assure that their resources are stretched as far as possible.

The primary source of retirement income is also expected to be related to whether an individual has a well-articulated retirement plan. The use of savings or personal investments as the primary source of retirement income is likely to require greater involvement of the individual than other sources of funding. These individuals are, therefore, expected to be more likely to have a clear retirement plan than other individuals. Insofar as defined-benefit and defined-contribution retirement accounts are arranged by employers, the establishment of such accounts may serve as a facilitating step in the process of thinking and planning for retirement. Defined-contribution plans require the individual to take greater responsibility for his/her retirement income than defined-benefit plans do. Individuals whose primary source of retirement income is a defined-contribution plan are, therefore, expected to be more likely to have clear retirement plans than those whose primary source is a defined-benefit plan. Those who rely primarily on Social Security are expected to be least likely to have a well-articulated retirement plan. For many individuals who are relying only on Social Security for retirement there is little incentive to plan; Social Security is what it is. The only planning decision is whether to take the Social Security benefits at age 62 or wait.

It should be noted that many individuals may have multiple sources of retirement savings. However, some of these sources may provide very little income, and it is expected that the primary source would be most indicative of the nature of an individual's involvement in retirement planning.

Whether an individual has discussed retirement planning with a professional financial planner is also expected to affect the probability of having a well-articulated retirement plan. For many individuals, working with a professional financial planner may be the most effective means of obtaining relevant financial knowledge. A planner can review the individual's current situation and suggest how best to approach saving for retirement. In most cases, recommendations would include development of a retirement plan with specific financial goals.

Several other variables may also be related to retirement planning. Education in general may be correlated with financial literacy, which is expected to aid the planning process. Financial literacy enables the individual to better understand important issues such as the relation between the value of current and future assets. Self-employment may also play a role. By taking responsibility for their own business, the self-employed demonstrate initiative and interest in business issues. They may, therefore, be more likely than other workers to plan for their retirement. Unions may serve as a source of information and support in the development of retirement plans. Individuals with poor health may see health care issues as an important consideration in their future and may consequently feel a greater need to plan.

Gender and marital status may also influence retirement planning. Workers who are married have been found to be more likely than workers who are not married to have put aside money for retirement and to have confidence in having enough money for a comfortable retirement (Helman et al., 2009). Women are less likely to have saved for retirement (Employee Benefit Research Institute and Mathew Greenwald & Associates, 2008), and they are less confident than men that they have enough money for retirement (Helman et al., 2009). The gender gap may be due partly to differences in income (Employee Benefit Research Institute and Mathew Greenwald & Associates, 2008). Gender differences in risk aversion may also affect savings behavior (Jianakoplos & Bernasek, 1998; Fisher, 2010).

DATA

Data in the current study were collected using the third Widener Elder Pennsylvanian Survey (WEPS), which was funded by the Pennsylvania Department of Education. These surveys used random calls to gather data from samples of 750 Pennsylvanians born in 1964 or earlier; thus, the data provide insights on how baby boomers and the elderly perceive their current and future lives. Survey questions were developed by faculty specializing in research in health and in economics with input from Mathew Greenwald and Associates, a public opinion and market research firm focusing on the financial services and retirement industries. The third survey (WEPS3), which was conducted in October 2008, was limited to individuals who were currently working. Issues of health, wealth, and work were explored, and the data provide a basis for analyzing expected retirement patterns—especially for the boomers.

In WEPS3, respondents were asked if they had developed a financial plan with specific retirement goals. In the current analysis, individuals who had such plans were labeled planners, while those who did not were labeled nonplanners. Respondents also indicated the source that was expected to account for the largest amount of their retirement income: (a) savings or income from personal investments; (b) employer-funded pension plan; (c) 401(k), IRA, or other retirement plan; or (d) Social Security benefits. In addition, respondents were also asked about their physical health and whether they were self-employed. Information on birth year, gender, marital status, education, union membership, and financial net worth were gathered as well. There were 509 observations with complete information on all the variables used in the analysis. Definitions of the variables and descriptive statistics are summarized in Table 1.

TABLE 1 Variable Definitions and Summary Statistics (n = 509)

Variable	Definition	Mean	Std. Dev.
Retirement plan with goals	1 if have retirement plan with specific goals;0 otherwise	0.4361	0.4964
Age	Age in years	54.7289	7.6802
Female	1 if female;0 if male	0.5088	0.5004
Married	1 if married;0 otherwise	0.7151	0.4518
No college	1 if no college completed;0 otherwise	0.3242	0.4685
Some college	1 if completed some college;0 otherwise	0.2613	0.4398
College graduate	1 if completed a bachelor's degree or more;0 otherwise	0.4145	0.4931
High financial net worth	1 if net worth ≥500,000;0 otherwise	0.2259	0.4186
Middle financial net worth	1 if net worth is ≥100,000 and <500,000;0 otherwise	0.3536	0.4786
Low financial net worth	1 if net worth is <100,000;0 otherwise	0.4204	0.4941
Self-employed	1 if self-employed;0 otherwise	0.0963	0.2952
Union member	1 if union member;0 otherwise	0.1611	0.3680
Poor health	1 if self-identified physical health is “poor”;0 if self-identified physical health is “excellent,” “good,” or “fair”	0.0275	0.1637
Defined benefit account is largest source of expected retirement income	1 if true;0 otherwise	0.2652	0.4419
Defined contribution account is largest source of expected retirement income	1 if true;0 otherwise	0.3615	0.4809
Savings or income from personal investments is largest source of expected retirement income	1 if true;0 otherwise	0.1827	0.3868
Social Security is largest source of expected retirement income	1 if true;0 otherwise	0.1906	0.3931
Spoke to professional planner	1 if discussed retirement planning with a professional financial planner0 otherwise	0.5187	0.5001

The data revealed that 44% of respondents had a retirement plan with specific goals. Furthermore, among the various sources of income for retirement, defined-contribution accounts were found to be particularly prevalent. Eighty-nine percent of respondents indicated that they or their spouse had savings in this type of account (not shown). (Note. The data do not indicate if a defined-benefit or defined-contribution plan was derived from the husband's or wife's employment for married households.) However, only for 36% of respondents were those accounts expected to be the primary source of retirement income. Sixty percent of respondents said they or their spouse participated in a defined-benefit program; 27% anticipated that this program would be their largest source of retirement income. Also, 60% of respondents had put aside other savings for retirement, but only 18% expected those savings to be their primary income source. While almost all workers participate in the Social Security program, just 19% anticipate that Social Security benefits would be their main income source. (The data do not provide the dollar amount of income expected to be provided by each source. That type of information can be difficult to obtain. For example, 18% of the respondents for the second Widener Elder Pennsylvanian Survey did not know their expected yearly Social Security income, and an additional 10% refused to answer the question, [Brucker & Leppel, 2008]).

DIFFERENCES BETWEEN PLANNERS AND NONPLANNERS

To determine which individuals are more likely to have a retirement plan with specific goals, the characteristics of planners and nonplanners were compared and statistical tests were performed. For continuous variables, t tests on the difference in means were calculated. For categorical variables, Z tests on the differences in proportions were computed. (See Table 2.)

TABLE 2 Comparison of Planners and Nonplanners

	Planners (n = 222)		Nonplanners (n = 287)		p-value of test of difference in means or proportions ^a
Variable	Mean	Std. Dev.	Mean	Std. Dev.
Age	55.1441	7.6901	54.4077	7.6705	0.2846
Female	0.4640	0.4998	0.5436	0.4990	0.0749
Married	0.7523	0.4327	0.6864	0.4648	0.1027
No college	0.2973	0.4581	0.3449	0.4762	0.2547
Some college	0.2523	0.4353	0.2683	0.4438	0.6829
College graduate	0.4505	0.4987	0.3868	0.4879	0.1481
High financial net worth	0.3378	0.4740	0.1394	0.3469	<.0001
Middle financial net worth	0.3739	0.4849	0.3380	0.4738	0.4009
Low financial net worth	0.2883	0.4540	0.5226	0.5004	<.0001
Self-employed	0.1351	0.3426	0.0662	0.2491	0.0089
Union member	0.1757	0.3814	0.1498	0.3575	0.4315
Poor health	0.0270	0.1625	0.0279	0.1649	0.9538
Defined benefit account is largest source of expected retirement income	0.2387	0.4273	0.2857	0.4525	0.2338
Defined contribution account is largest source of expected retirement income	0.4234	0.4952	0.3136	0.4648	0.0105
Savings or income from personal investments is largest source of expected retirement income	0.2613	0.4403	0.1220	0.3278	0.0001
Social Security is largest source of expected retirement income	0.0766	0.2665	0.2787	0.4492	<.0001
Spoke to professional planner	0.7387	0.4403	0.3484	0.4773	<.0001

First, planners were found to be much more likely to have high net worth and much less likely to have low net worth. Furthermore, they were more likely to have savings or personal investments or a defined-contribution plan as their primary source of retirement income. They were less likely to have Social Security as their primary source. In addition, a gender gap in retirement planning was evident, with women being less likely to have a well-articulated plan. Moreover, the self-employed were more heavily represented among the planners. Planners were also more likely to have discussed retirement with a professional financial planner.

Surprisingly, there were no significant differences in education for planners versus nonplanners. It is possible that differences in education in general are not important factors here. Rather, differences in financial literacy in particular may be what matters. A related finding by Fisher (2010) indicated that while education did influence the likelihood of men regularly setting aside money for saving, education had no significant effect on women's saving.

There was also no significant difference between planners and nonplanners in the percentage with poor health; this may be due to the small amount of variation in the health variable. Since all the respondents reported that they were working and not retired, it is likely that they all had reasonably good health. Eighty-six percent of respondents identified their health as excellent or good.

The results of this analysis should be interpreted with caution in light of the study's limitations. Since all the survey participants are residents of Pennsylvania, the results may not generalize across states or countries. Furthermore, since approximately 93% of the sample is Caucasian, the results may not apply to minority populations. Despite these limitations, the study has taken a step in the direction of improved understanding of retirement planning issues.

CONCLUSIONS

This analysis has identified numerous differences in the characteristics of those who have a retirement plan with specific goals (planners) and those who do not (nonplanners). The finding of different characteristics of planners and nonplanners is consistent with the conclusions of MacFarland, Marconi, and Utkus (2004), which indicated that individuals differ in their tastes for financial planning activities.

Characteristics such as having low net worth and expecting to rely primarily on Social Security do not bode well for the retirement preparedness of nonplanners. Such attributes suggest that many nonplanners are likely to be at risk of a reduction in their standard of living after retirement. Individuals whose largest source of retirement saving is Social Security, in many cases, may be people whose employers do not offer retirement plans. Without this facilitating factor, the process of planning for retirement is likely to be much more difficult. Approximately 30% of employees work for an organization that does not offer a retirement plan (Sanzenbacher, 2006). Greater access to employer-provided retirement plans would aid retirement preparedness.

As Selnow (2004) indicated in his discussion of costs and benefits, there is no immediate reward for retirement planning and saving and no immediate penalty for not doing it. In addition, the power of inertia makes it easy to become overwhelmed by the prospect of financial planning and, consequently, to do nothing about it. If the perceived immediate costs and benefits can be altered, financial planning is more likely to occur. Baby boomers have frequently used computers at work and are often comfortable with the notion of technology, particularly if it is user-friendly. The cohorts that follow the baby boomers are even more comfortable with technology and have used computers in their leisure activities as well as at work. That comfort with technology can be used to advantage in financial planning. User-friendly software with an aesthetically appealing interface can be developed to make financial planning easy and even fun. Making the process easier reduces the current costs of financial planning and making it fun increases the current benefits; thus, financial planning becomes more likely. As Ajzen's (1991) model implies, retirement planning is less likely for those who see a plan as having minimal chance of future success such as low income individuals and those expecting to be primarily reliant on Social Security. For these individuals, it is even more important to alter the current costs and benefits. If they can be enticed to play with planning software because it is fun, they can then see how much a small amount of saving can grow and they may be motivated to start the financial planning and saving process.

Trends and patterns with respect to gender and computing have implications for the effectiveness of user-friendly software for increasing women's level of retirement planning. While historically men were more inclined to use the Internet, the evidence indicates that the gender gap has closed (Fallows, 2005). Differences in how the Internet is used, however, continue. Women have been found to be more prone than men to use the Internet for interpersonal communication and education assistance (Weiser, 2000). That inclination to use the Internet for educational purposes suggests that well-designed software can be an effective way to enhance women's retirement planning.

Improved awareness of financial issues would also promote greater retirement planning and saving. Workers who have received employer-provided financial education have been found to be more likely to have a retirement savings program (Joo & Grable, 2005). Furthermore, individuals who exhibit greater financial literacy are more likely to be successful planners (Lusardi & Mitchell, 2006, 2008). Thus, clarification of workers’ retirement goals and general knowledge of the savings requirements necessary to meet those goals would help to provide the impetus to participate in retirement savings programs. After that, programs such as the Save More Tomorrow^TM plan can use inertia to finish the job.

To conclude, the trend away from defined-benefit accounts and toward defined-contribution accounts requires that individuals take greater responsibility for retirement planning. However, many workers are not predisposed to be planners; procrastination, inertia, and low expectations of success often prevent them from taking the necessary steps to attain their financial goals. Three general recommendations are therefore provided: (a) develop user-friendly software to entice individuals to explore retirement planning in a fun way; (b) improve financial literacy education to provide additional understanding and motivation; and (c) institute programs that use inertia to encourage greater saving. Grants by public and private institutions can be used to promote the development of appropriate software. When fun, game-like software has been developed, firms can run company competitions with small prizes to encourage employees to play with the software. In addition, access to professional financial planners and financial literacy training can be provided partly by firms. Local governments can also offer subsidized adult education courses in financial literacy and promote attendance at retirement seminars. We are living in an age when basic arithmetic skills are no longer sufficient. Individuals must be able to make informed decisions to appropriately manage their personal finances. It is also recommended that public service advertising be used to encourage individuals to supplement their anticipated Social Security benefits with other sources of retirement income. It is hoped that the current political debate over the future of Social Security will generate increased interest not just in the future of Social Security but also in more comprehensive financial planning for retirement.

Notes

^a The test statistic is for the difference in means for age. For other variables, the test is for the difference in proportions.