Issue Brief. Employment-Based Retirement and Pension Plan Participation: Geographic Differences and Trends EBRI EMPLOYEE BENEFIT RESEARCH INSTITUTE

Transcription

1 April 2003 Jan Employment-Based Retirement and Pension Plan Participation: Geographic Differences and Trends by Craig Copeland, EBRI Feb Mar Apr May Jun Jul Aug EBRI EMPLOYEE BENEFIT RESEARCH INSTITUTE Issue Brief This Issue Brief examines the level of participation by workers in employment-based pension or retirement plans. Their participation is examined for 2001 across various worker characteristics and those of their employers. The report then examines retirement plan participation across U.S. geographic regions, including by state and certain consolidated metropolitan statistical areas (CMSAs). Participation trends from 1987 to 2001 are also presented. Of all million workers in 2001, 83.5 million worked for an employer or a union that sponsored a pension or retirement plan, and 64.9 million participated in the plan. This translates into a sponsorship rate (the percentage of workers working for an employer or a union that sponsored a plan) of 55.3 percent and a participation level among all workers of 43.0 percent. For wage and salary workers ages 21 64, the sponsorship rate increases to 61.7 percent and the portion participating increases to 49.8 percent. The bigger the employer, the more likely a worker is to participate in an employmentbased retirement plan. Among workers who worked for employers with fewer than 10 employees, 18.0 percent participated in a plan, compared with 61.6 percent of those working for an employer with 1,000 or more employees. The states with the lowest levels of participation Florida, Texas, California, and Washington state were in the South, Southwest, and the upper Northwest. The states with the highest participation were in the upper Midwest Minnesota, Wisconsin, Michigan, and the Dakotas along with some in the Northeast or Mid-Atlantic, such as Pennsylvania and Maryland. For the all worker population, the percentage of workers participating in an employment-based retirement plan increased from 37.6 percent in 1987 to 39.7 percent in 1991, then fell to 39.0 percent in The percentage subsequently increased steadily through 2000 to 44.4 percent, before declining significantly in 2001 to 43.0 percent. The all wage and salary workers ages and private-sector wage and salary workers ages populations mirrored this pattern. In contrast, full-time, full-year wage and salary workers ages had an almost constant level of retirement plan participation over the period, with a minor increase in the late 1990s. Public-sector wage and salary workers had a pattern similar to that of the full-time, full-year workers. Small employer, part-time, and part-year workers had higher retirement plan participation levels in 2001 relative to 1987, which is good news. Nevertheless, as the economy has remained stagnant, these workers are the ones less likely to be able to participate, as employers may have to cut back on their offerings to remain profitable in a weaker economy. The bad news is that the workers with traditionally high levels of participation did not have a significant increase in their level of participation during a very prosperous time in the American economy in the 1990s. Thus, the overall level of participation in retirement plans has shown only a small increase from 1987 to 2001 and appears, in the short term, to be headed downward. Sep Oct Nov Dec 2003 EBRI Issue Brief Number 256 April EBRI

2 Table of Contents Retirement Plan Types... 4 Data Levels... 6 Worker Characteristics... 6 Employer Characteristics... 8 A Closer Examination... 8 Geographic Differences Trends Trends in the Percentage Participating Trends Across Worker Characteristics Conclusion References Endnotes Figures Figure 1, Percentage of Various Work Forces Who Work for an Employer That Sponsored a Retirement Plan and the Percentage Who Participated in the Plan, Figure 2, Percentage of Various Work Forces Who Worked for an Employer That Sponsored a Retirement Plan and the Percentage That Participated in the Plan, by Various Characteristics, Figure 3, Percentage of Wage and Salary Workers Ages Who Participated in an Employment-Based Retirement Plan, by Work Status and Gender, Figure 4, Percentage of Wage and Salary Workers Ages Who Participated in an Employment-Based Retirement Plan, by Annual Earnings and Gender, Figure 5, Percentage of Wage and Salary Workers Ages Who Participated in an Employment-Based Retirement Plan, by Annual Earnings and Race/Ethnicity, Figure 6, Percentage of Wage and Salary Workers Ages Who Participated in an Employment-Based Retirement Plan, by Age and Race/Ethnicity, Figure 7, Percentage of Wage and Salary Workers Ages Who Participated in an Employment-Based Retirement Plan, by Firm Size (Number of Employees) and Race/Ethnicity, Figure 8, Percentage of Wage and Salary Workers Ages Who Participated in an Employment-Based Retirement Plan, by Age and Employer Size (Number of Employees), Figure 9, Percentage of Wage and Salary Workers Ages Who Participated in an Employment-Based Retirement Plan, by Annual Earnings and Employer Size (Number of Employees), Figure 10, Percentage of Various Work Forces Who Participated in an Employment-Based Retirement Plan, by State, Figure 11, Percentage of Wage and Salary Workers Who Participated in an Employment-Based Retirement Plan by State, Figure 12, Percentage of Various Work Forces Who Participated in an Employment-Based Retirement Plan, by Consolidated Metropolitan Statistical Area (CMSA), Figure 13, Percentage of Various Work Forces Who Worked for an Employer That Sponsored a Retirement Plan and the Percentage Who Participated in an Employment-Based Retirement Plan, Figure 14, Percentage of Various Work Forces Who Participated in an Employment-Based Retirement Plan, Figure 15, Percentage of Wage and Salary Workers Ages Who Participated in an Employment-Based Retirement Plan, by Age, Figure 16, Percentage of Wage and Salary Workers Ages Who Participated in an Employment-Based Retirement Plan, by Gender, April 2003 EBRI Issue Brief

3 Figure 17, Percentage of Wage and Salary Workers Ages Who Participated in an Employment-Based Retirement Plan, by Race/Ethnicity, Figure 18, Percentage of Wage and Salary Workers Ages Who Participated in an Employment-Based Retirement Plan, by Educational Level, Figure 19, Percentage of Wage and Salary Workers Ages Who Participated in an Employment-Based Retirement Plan, by Earnings ($ 2001), Figure 20, Percentage of Wage and Salary Workers Ages Who Participated in an Employment-Based Retirement Plan, by Work Status, Figure 21, Percentage of Wage and Salary Workers Ages Who Participated in an Employment-Based Retirement Plan, by Workers Employer Size, Figure 22, Percentage of Wage and Salary Workers Ages Who Participated in an Employment-Based Retirement Plan, by Industry/Sector, Craig Copeland of EBRI wrote this Issue Brief with assistance from the Institute s research and editorial staffs. Any views expressed in this report are those of the author and should not be ascribed to the officers, trustees, or other sponsors of EBRI, EBRI- ERF, or their staffs. Neither EBRI nor EBRI-ERF lobbies or takes positions on specific policy proposals. EBRI invites comment on this research. This publication is available for purchase online. Visit or call (202) As the United States population Introduction has aged due to increases in longevity and the aging of the post- World War II baby-boom generation, financial preparation for the retirement of this and future generations has become a national concern. This concern has arisen for various reasons, including the projected deficit in the Social Security system, continual increases in health care costs, and the significant amount of resources this growth in the aged population will require relative to the smaller working population supporting it. A primary source of funds for many retirees is a pension or retirement plan offered through their former employer or union. The larger the percentage of retirees who have a pension or retirement plan, the smaller the burden will likely be on future generations to provide this support, as retirees should be better able to pay for their own expenses in retirement from these plans. Consequently, tracking the percentage of the working population with a pension or retirement plan gives legislative decision makers, employers, and workers an idea of the extent to which the potential burden on tax payers of supporting future retirees may be changed as a result of these trends in the levels of workers participating in retirement plans. This Issue Brief examines the level of participation by workers in employment-based pension or retirement plans. It begins with a breakdown of retirement plan types and participation in these plan types. Next, it describes the data used in this study, along with their relative strengths and weaknesses. From these data, results on participation in employment-based retirement plans are examined for 2001 across various worker characteristics and those of their employers. The report then examines retirement plan participation across U.S. geographic regions, including a breakdown by state and certain consolidated metropolitan statistical areas (CMSAs). In addition to the results for 2001, trends April 2003 EBRI Issue Brief 3

4 from in employment-based retirement plan participation are presented across the same worker characteristics and those of their employers as used for The report concludes with a discussion of this study s findings. Retirement Plan Types In general, there are two types of tax-favored retirement plan vehicles offered through employers and unions defined contribution (DC) plans and defined benefit (DB) plns. These plans differ in variety of ways, particularly in how contributions are made, the assumption of investment risk, and the manner in which plan benefits are paid. First, under a DC plan, employer contributions (if any) are based on a predetermined formula, 1 and, most frequently, all contributions (made by both employers and/or employees) are placed in individual accounts on behalf of each participant. In contrast, DB plans typically are funded by the employer and do not require participants to contribute, and plan contributions are held in one trust on behalf of all participants, with these contributions subject to federal funding rules and regulations required to maintain the plan s tax-favored status. Next, who assumes the investment risk associated with plan assets is a key distinction between DC and DB plans. The overwhelming majority of individuals receiving DC plan benefits assume all of the investment risk on their own accounts; that is, employers do not guarantee a specific benefit level but instead the benefit is determined by the contributions (employer and employee) to the individual s account and the investment returns within that account. 2 In comparison, DB plan participants receive a certain benefit amount calculated from a specific formula, typically based on average salary and years of service, regardless of the investment performance of the plan assets. Thus, in general, individual participants bear the investment risk in DC plans, while the plan sponsor does in DB plans. A third difference between DC and DB plans traditionally has been the manner in which they generally pay benefits. DC plans usually pay out benefits in a lump sum the entire accumulated benefits paid out at one time. Consequently, the recipients are responsible for managing the money to reduce or eliminate the risk of outliving the assets during their retirement. Alternatively, DB plans must offer life annuities (a set amount paid out regularly over time, typically monthly, for as long as the beneficiary lives), which, when chosen, eliminates the necessity of managing these assets during retirement. However, plan sponsors are allowed to cash out those participants who terminate employment and have a small accrued benefit, and a growing number of DB plan participants are also being offered a lump-sum distribution option. 3 The term pension plan traditionally has been synonymous with a DB plan that uses a fixed annuity payment, and not with a DC plan offering a lump-sum distribution. Although many individuals refer to a DC plan as a pension plan, many other people still understand a pension to be an annuity payment at retirement (i.e., a DB plan). Thus, this discussion defines DB and DC plans as retirement plans to eliminate any confusion. The increase in the number of DC plan participants relative to DB plan participants has been well-documented. For example, Rajnes (2002), using results from Internal Revenue Service Form 5500 data, shows that the number of private-sector workers with a DB plan as their primary retirement plan decreased from 27 million in 1975 to 23 million in 1998 (a decline of almost 15 percent), while those with a DC plan as their primary plan increased from 4 million in 1975 to 29 million in 1998 (an increase of 625 percent over the same period). Copeland (2002b), using the U.S. Census Bureau s May 1988 Current Population Survey Employee Benefit Supplement, and the Retirement and Pension Plan Coverage Topical Module to the 1996 Survey of Income 4 April 2003 EBRI Issue Brief

5 and Program Participation (SIPP), found that the percentage of all civilian nonagricultural wage and salary workers age 16 or older who considered their primary retirement to be a DC plan increased from 25.8 percent in 1988 to 62.7 percent in Another study by Copeland and VanDerhei (2000), using the 1992 and the 1998 Survey of Consumer Finances (SCF), found that the percentage of families with at least one participant in a retirement plan and with only a DC plan increased from 38 percent in 1992 to 57 percent in The reasons for this growth of DC plan participants, such as the perception of increased work force mobility and changes in the business and regulatory environments of plan sponsors, have also been documented. 4 The consequences of these trends for retirees are significant, the most important being the necessity for individual retirees to manage their assets in retirement so as not to outlive them. However, these issues are outside the scope of this report, which focuses on the initial accumulation of retirement benefits through participation in an employment-based arrangement. 5 While the breakdown of employment-based Data retirement plan participation among plan types is important, the datasets that contain this information are slow to be released, or are compiled from surveys taken only at three- or five-year intervals. The data cited above on the breakdown of plan types from the Survey of Consumer Finances (SCF) and Survey of Income and Program Participation (SIPP) illustrate this problem, as the latest available numbers are from 1998, with the survey intervals being three years and five years, respectively. 6 Furthermore, the official compilation of private-sector plan assets and participants by the U.S. Department of Labor from the Form 5500 data that all private-sector sponsors of pension or retirement plans must file with the Internal Revenue Service (IRS) is currently only available through Public-sector employers are not required to file these forms, making detailed data from those plans difficult to obtain even after a few years. Consequently, timely data on the breakdown between retirement plan types is not available on an annual basis. In contrast, the Current Population Survey (CPS), a monthly survey, asks questions in a consistent manner each March since at least 1988 about whether a worker worked for an employer or union that sponsored a pension or retirement plan for any of its employees, and then if the worker was included in that plan. 7 The U.S. Census Bureau conducts the CPS by interviewing more than 60,000 households and asking numerous questions about individuals work status, employers, income, and basic demographic characteristics. Therefore, the CPS provides detailed information about workers from a broad sample of Americans who are in an employment-based plan, making it possible to establish a consistent annual and timely trend across numerous worker characteristics and the characteristics of their employers. While the CPS provides excellent detail on overall participation in employment-based plans, it does not provide specifics about the plans such as the worker s plan type or whether the worker is eligible to participate in the plan sponsored by his or her employer or union. This makes the definition of terms in this study important. The term sponsorship rate is defined as the percentage of workers in the specified work force who worked for an employer or union that sponsored a plan for any of its employees, not necessarily for the worker in question. Thus, in this discussion, the term percentage of workers participating in a plan is not synonymous with the standard retirement plan term participation rate, which is understood to mean the percentage of eligible workers who participate in a plan. Consequently, participation rate is not used in this analysis; instead the terms participation level or percentage participating, which, to reiterate, refer to the fraction of workers in the specified work force who participated in an employmentbased pension or retirement plan regardless of the April 2003 EBRI Issue Brief 5

6 Figure 1 Percentage of Various Work Forces Who Work for an Employer That Sponsored a Retirement Plan and the Percentage Who Participated in the Plan, 2001 Full-Time, Wage Private Wage Public Wage Full-Year and Salary and Salary and Salary Wage and Workers Workers Workers Salary Workers All Workers Ages Ages Ages Ages (millions) Worker Category Total Works for an employer sponsoring a plan Participating in a plan (percentage) Worker Category Total 100.0% 100.0% 100.0% 100.0% 100.0% Works for an employer sponsoring a plan Participating in a plan Source: Employee Benefit Research Institute estimates from the March Current Population Surveys worker s eligibility (offered a plan and meets the requirements to participate) to participate in a plan. Lastly, the term participating in a plan as used here always means a pension or retirement plan provided through an employment-based arrangement, not a plan such as an individual retirement account (IRA) that workers can fund outside of an employment-based arrangement. Of all million workers in 2001, Levels million worked for an employer or union that sponsored a pension or retirement plan, and 64.9 million participated in the plan (Figure 1). This translates into a sponsorship rate (the percentage of workers working for an employer or union that sponsored a plan) of 55.3 percent and a participation level of 43.0 percent. However, this measure of the work force contains the self-employed and those typically with a looser connection to the work force individuals under age 21 and older than age 64. Therefore, a different measure of the work force, wage and salary workers ages representing individuals who have a stronger connection to the work force and work for someone else is examined. For this group, the sponsorship rate increases to 61.7 percent and the portion participating increases to 49.8 percent. When separating these wage and salary workers into the public and private sectors, the percentages participating differ significantly. Slightly more than 75 percent of the public-sector workers participated in an employment-based retirement plan, compared with 45.1 percent of the private-sector workers. A more restrictive definition of the work force, which more closely resembles the types of workers who generally must be covered in accordance with the Employee Retirement Income Security Act (ERISA) for a retirement plan offered by a private-sector employer or union, is the work force of full-time, full-year wage and salary workers ages Approximately 58 percent of these workers participated in a retirement plan. The percentage participating in each work force definition is correlated with the workers characteristics as well as those of the workers employers. The remainder of this section focuses on wage and salary workers, presenting the differences across these characteristics, which, in general, were representative of all the work force populations, except where noted. Worker Characteristics The percentage of wage and salary workers ages participating in a retirement plan increased through age 55 then declined (Figure 2). For those ages 21 24, 20.3 percent participated in a plan, compared with nearly 60 percent of those ages Males were more likely to participate in a plan than females. However, there were no differences between males and females among full-time, full-year workers. Being white or having attained a higher educational level was associated with a higher probability of participating in a retirement plan. Workers who were married or divorced were more likely to participate in a plan, while never-married workers had the lowest probability. The higher an 6 April 2003 EBRI Issue Brief

7 Figure 2 Percentage of Various Work Forces Who Worked for an Employer That Sponsored a Retirement Plan and the Percentage Who Participated in the Plan, by Various Characteristics, 2001 Full-Time, Full-Year Wage and Salary Private Wage and Salary Public Wage and Salary Wage and Salary All Workers Workers Ages Workers Ages Workers Ages Workers Ages Number Sponsor- Percent- Number Sponsor- Percent- Number Sponsor- Percent- Number Sponsor- Percent- Number Sponsor- Percentof ship age of ship age of ship age of ship age of ship age workers rate participating workers rate participating workers rate Participating workers rate participating workers rate participating (000s) (%) (%) (000s) (%) (%) (000s) (%) (%) (000s) (%) (%) (000s) (%) (%) Age 20 or younger 12, % 4.8% , , % 20.3% 10, % 19.1% 1, % 32.4% 5, % 30.7% , , , , , , , , , , , , , , , , , , , , or older 5, Gender Male 79, , , , , Female 71, , , , , Race/Ethnicity White 108, , , , , Black 16, , , , , Hispanic 17, , , , , Other 7, , , , Education No high school diploma 20, , , , High school diploma 45, , , , , Some college 42, , , , , Bachelor s degree 27, , , , , Graduate/profnl. degree 13, , , , , Marital Status Married 85, , , , , Widowed 3, , , , Divorced 15, , , , , Separated 3, , , , Never married 43, , , , , Heath Insurance Status None/not own employer 68, , , , , Through own employer 82, , , , , Health Status Excellent 49, , , , , Very good 53, , , , , Good 37, , , , , Fair 8, , , , , Poor 1, , , Annual Earnings <$5,000 19, , , , $5,000 $9,999 13, , , , , $10,000 $14,999 14, , , , , $15,000 $19,999 13, , , , , $20,000 $29,999 26, , , , , $30,000 $39,999 21, , , , , $40,000 $49,999 13, , , , , $50,000 or more 28, , , , , Work Status Full-time, full-year 99, , , , Full-time, part-year 21, , , , Part-time, full-year 14, , , , Part-time, part-year 15, , , , Employer Size <10 employees 29, , , , employees 14, , , , employees 18, , , , employees 17, , , , employees 6, , , , ,000 or more employees 43, , , , Public sector 21, , , Sector/Industry Private sector 129, , , , agriculture, mining, and construction 14, , , , manufacturing 29, , , , wholesale/retail trade 51, , , , personal services 34, , , , Public sector 21, , , Source: Employee Benefit Research Institute estimates from the 2002 March Current Population Survey. April 2003 EBRI Issue Brief 7

8 70% Figure 3 Percentage of Wage and Salary Workers Ages Who Participated in an Employment-Based Retirement Plan, by Work Status and Gender, % 58.5% Male Female 32.8% 35.9% 27.4% 20.1% 12.0% 16.9% 10% 0% Full-Time, Full-Year Worker Full-Time, Part-Year Worker Part-Time, Full-Year Worker Part-Time, Part-Year Worker Work Status Source: Employee Benefit Research Institute estimates from 2002 March Current Population Survey. individual s earnings were, the more likely he or she participated in a plan. Approximately one-third of those who earned $15,000 $19,999 participated in a plan, while this number increased to 75.5 percent of those earning $50,000 or more. Health status and health insurance ownership also were associated with the probability of participating in a retirement plan. Workers with health insurance through their own employer were more than three times as likely to participate in a plan as those without health insurance through their own employer. Furthermore, as the workers self-reported health status decreased, so did their probability of participating in a retirement plan. Lastly, full-time, full-year workers were by far the most likely type to participate in a retirement plan. Employer Characteristics The probability of a worker participating in an employment-based retirement plan increased significantly with the size of his or her employer (Figure 2). For workers who worked for employers with fewer than 10 employees, 18.0 percent participated in a plan, compared with 61.6 percent of those working for an employer with 1,000 or more employees. The sector and industry of the employer also had an impact on the likelihood of participating in a plan. Public-sector workers were significantly more likely to participate than privatesector workers. Workers in the manufacturing industry had the highest probability of participating, while those in the agriculture, mining, and construction industry had the lowest probability. A Closer Examination Gender Female wage and salary workers ages were found to participate in a retirement plan at a lower level than males did. However, among full-time, full-year workers of the same ages, females had a slightly higher rate of participation in a plan (58.5 percent for women, compared with 58.1 percent for men). In fact, across all of the worker status categories, females were more likely to participate in a retirement plan than males (Figure 3). Furthermore, when examining the participation by earnings level, except for the lowest earnings category, the level of participation by females was also higher than that it was for males (Figure 4). Consequently, it appears that females lower probability of participation at the all-worker level was a result of their lower earnings and/or lower rates of full-time work in comparison with males. Race/Ethnicity Analysis of race/ethnicities by earnings level shows that both black and Hispanic wage and salary workers were significantly less likely than whites to participate in a retirement plan. However, the 8 April 2003 EBRI Issue Brief

9 Figure 4 Percentage of Wage and Salary Workers Ages Who Participated in an Employment-Based Retirement Plan, by Annual Earnings and Gender, % 80% Male 74.8% 77.3% 74.8% 70% Female 68.0% 68.5% 54.2% 58.3% 38.1% 42.1% 26.4% 28.3% 10% 10.8% 9.2% 16.6% 12.4% 18.3% 0% Less than $5,000 $5,000 $9,999 $10,000 $14,999 $15,000 $19,999 $20,000 $29,999 $30,000 $39,999 $40,000 $49,999 $50,000 or more Annual Earnings Source: Employee Benefit Research Institute estimates from 2002 March Current Population Survey. gap between the percentage of black and white plan participants narrowed at the higher earning levels, with black workers earning of $30,000 $39,999 having a higher level of participation (Figure 5). In contrast, the gap between Hispanics and whites persisted even in the upper earnings groups, although it showed some narrowing in the highest earnings group. Similarly, at ages 25 and older, white workers had significantly higher levels of retirement Figure 5 Percentage of Wage and Salary Workers Ages Who Participated in an Employment-Based Retirement Plan, by Annual Earnings and Race/Ethnicity, % 80% 70% White Black Other Hispanic 52% 51% 65% 65% 54% 47% 73% 72% 61% 57% 77% 75% 70% 66% 27% 38% 36% 31% 33% 10% 11% 9% 7% 5% 17% 15% 13% 9% 21% 21% 13% 19% 0% Less than $5,000 $5,000 $9,999 $10,000 $14,999 $15,000 $19,999 $20,000 $29,999 $30,000 $39,999 $40,000 $49,999 $50,000 or more Annual Earnings Source: Employee Benefit Research Institute estimates from 2002 March Current Population Survey. April 2003 EBRI Issue Brief 9

10 Figure 6 Percentage of Wage and Salary Workers Ages Who Participated in an Employment-Based Retirement Plan, by Age and Race/Ethnicity, % White 63% Black Other Hispanic 49% 58% 52% 48% 53% 55% 56% 58% 53% 55% 54% Hispanic-Native Born Hispanic-Non-Native Born 41% 41% 36% 35% 39% 39% 23% 22% 17% 15% 24% 16% 26% 28% 31% 10% 10% 0% Age Source: Employee Benefit Research Institute estimates from 2002 March Current Population Survey. Figure 7 Percentage of Wage and Salary Workers Ages Who Participated in an Employment-Based Retirement Plan, by Firm Size (Number of Employees) and Race/Ethnicity, % 80% White Black 78% 70% Other Hispanic 57% 61% 55% 66% 52% 70% 67% 62% 32% 45% 32% 33% 42% 41% 29% 43% 43% 10% 21% 15% 13% 7% 21% 19% 10% 22% 0% Fewer Than ,000 or more Public Firm Size (employees) Source: Employee Benefit Research Institute estimates from 2002 March Current Population Survey. 10 April 2003 EBRI Issue Brief

11 Figure 8 Percentage of Wage and Salary Workers Ages Who Participated in an Employment-Based Retirement Plan, by Age and Employer Size (Number of Employees), % 80% 70% Fewer than ,000 or More Public 69% 55% 44% 83% 78% 72% 68% 66% 59% 56% 57% 49% 44% 79% 70% 63% 44% 32% 29% 26% 22% 17% 14% 23% 33% 24% 34% 31% 10% 7% 9% 0% Age Source: Employee Benefit Research Institute estimates from 2002 March Current Population Survey. plan participation than the other groups, with black workers narrowing the gap somewhat at older ages while Hispanic workers participation still remained considerably below the other race/ethnicity levels (Figure 6). However, a dramatic shift occurred in the results for Hispanic worker participation levels when the Hispanic workers were divided by birthplace United States or outside the United States. Native-born Hispanic workers age 25 or older had participation levels very similar to those of black workers, but still lower than white workers. In contrast, non-native born Hispanic workers had substantially lower levels of participation across all age groups. Another potential contributor to the lower rate of participation by Hispanic workers could be the characteristics of their employers, such as firm size (number of employees). However, across all the firm sizes, including public-sector employers, this appears not to be the case, as Hispanic workers had significantly lower participation in employment-based retirement plans than workers of other races/ethnicities (Figure 7). For workers at the smallest employers (fewer than 10 employees), 21 percent of white wage and salary workers participated in a plan, compared with 7 percent of the Hispanic workers. These levels increased as the employer size increased, but white workers still had a significantly higher participation level among those working for employers with 1,000 or more employees (66 percent for whites versus 43 percent for Hispanics). Consequently, while blacks with higher earnings had levels of retirement plan participation similar to those of whites, Hispanics had persistently lower levels of participation across earnings, age, and employer size. However, native-born Hispanic workers displayed participation levels similar to those of black workers, levels much closer to those of white workers. Firm Size Employees of firms with fewer employees were significantly less likely to participate in a retirement plan. Therefore, a potential explanation for this lower participation could be that these firms employed workers with characteristics associated with lower participation, such as being younger or lower paid. However, when controlling for age, workers at smaller employers still had a persistently lower level of participation across the age groups (Figure 8). Furthermore, across various earnings levels, workers of small employers were less likely to participate in an employment-based retirement plan (Figure 9). Even among workers making $50,000 or more, a considerable disparity exists 39 percent of those working for the smallest employers participated in a plan, compared with 85 percent of those working for employers with 1,000 or more employees. April 2003 EBRI Issue Brief 11

12 Figure 9 Percentage of Wage and Salary Workers Ages Who Participated in an Employment-Based Retirement Plan, by Annual Earnings and Employer Size (Number of Employees), % 90% 80% 70% Fewer than ,000 or More 74% 87% 75% 69% 63% 77% 73% 66% 85% 91% Public 22% 18% 13% 21% 37% 55% 45% 35% 23% 16% 53% 43% 29% 52% 39% 10% 5% 7% 0% Less than $15,000 $15,000 $29,999 $30,000 $49,999 $50,000 or more Annual Earnings Source: Employee Benefit Research Institute estimates from 2002 March Current Population Survey. Geographic Differences Not only did the workers characteristics affect the probability of their participation in an employment-based retirement plan, but their geographic location also had an impact. Wage and salary workers ages living in Florida had the lowest probability (40.6 percent) of participating in a plan in 2001, while those living in Minnesota had the highest probability (60.1 percent) (Figure 10). A similar pattern emerges for other workforce definitions, as workers in Minnesota again had the highest probability of participation among full-time, fullyear workers; private wage and salary workers; and all workers. Florida had the lowest probability of participation among all workers and full-time, full-year workers, while New Mexico s private wage and salary workers had the lowest likelihood of participation (Florida s private wage and salary workers had the next-lowest percentage). Public-sector wage and salary workers did not have the same states with the highest and lowest levels as the other work forces, as public-sector workers in Vermont had the lowest percentage of workers participating. Those workers in Pennsylvania had the highest percentage. Florida public-sector wage and salary workers were once more toward the bottom, while those in Minnesota were again near the top. 9 Using wage and salary workers ages as the work force proxy for the other work-force populations (with the exception of the public-sector workers) showed regional differences across the United States along with those among the states. The states with the lowest levels of participation e.g., Florida, Texas, California, and Washington state were in the South, Southwest, and the upper Northwest (Figure 11). The states with the highest participation were in the upper Midwest e.g., Minnesota, Wisconsin, Michigan, and the Dakotas along with some in the Northeast or Mid-Atlantic, such as Pennsylvania and Maryland. In general, the Midwestern states had the higher participation levels, while the Southern states had the lower levels. Certain consolidated metropolitan statistical areas (CMSAs) are identified in the CPS, and again those CMSAs located in the South and West e.g., Miami-Fort Lauderdale, FL, and Los Angles-Riverside-Orange County, CA had the lowest retirement plan participation levels for wage and salary workers (Figure 12). The highest participation levels for the wage and salary workers were for those workers in the Milwaukee-Racine, WI, CMSA. Public-sector wage and salary workers in these same CMSAs had the lowest and highest participation levels, respectively, with the exception of Detroit-Ann Arbor-Flint, MI replacing the Los Angles area as having the second lowest participation level. The Los Angles area did have the third-lowest level for the public-sector workers. 12 April 2003 EBRI Issue Brief

13 Figure 10 Percentage of Various Work Forces Who Participated in an Employment-Based Retirement Plan, by State, 2001 Full-Time, Full-Year Wage and Salary Private Wage and Salary Public Wage and Salary Wage and Salary All Workers Workers Ages Workers Ages Workers Ages Workers Ages Number Participating Number Participating Number Participating Number Participating Number Participating State (millions) (%) (millions) (%) (millions) (%) (millions) (%) (millions) (%) All % % % % % Alabama Alaska Arizona Arkansas California Colorado Connecticut Delaware District of Columbia Florida Georgia Hawaii Idaho Illinois Indiana Iowa Kansas Kentucky Louisiana Maine Maryland Massachusetts Michigan Minnesota Mississippi Missouri Montana Nebraska Nevada New Hampshire New Jersey New Mexico New York North Carolina North Dakota Ohio Oklahoma Oregon Pennsylvania Rhode Island South Carolina South Dakota Tennessee Texas Utah Vermont Virginia Washington West Virginia Wisconsin Wyoming Source: Employee Benefit Research Institute estimates from the 2002 March Current Population Survey. April 2003 EBRI Issue Brief 13

14 Figure 11 Percentage of Wage and Salary Workers Who Participated in an Employment-Based Retirement Plan by State, 2001 Percentage Distribution 55.0% to 60.2% (11) 53.1% to 55.0% (7) 50.4% to 53.1% (11) 46.7% to 50.4% (10) 40.6% to 46.7% (12) Source: Employee Benefit Research Institute estimates from the 2002 March Current Population Survey. Figure 12 Percentage of Various Work Forces Who Participated in an Employment-Based Retirement Plan, by Consolidated Metropolitan Statistical Area (CMSA), 2001 Full-Time, Full-Year Wage and Salary Private Wage and Salary Public Wage and Salary Wage and Salary All Workers Workers Ages Workers Ages Workers Ages Workers Ages Number Participating Number Participating Number Participating Number Participating NumberParticipating CMSA (millions) (%) (millions) (%) (millions) (%) (millions) (%) (millions) (%) All % % % % % Unrecorded CMSA Boston-Worcester-Lawrence, MA-NH-ME-CT Chicago-Gary-Kenosha, IL-IN-WI Cincinnati-Hamilton, OH-KY-IN Cleveland-Akron, OH Dallas-Fort Worth, TX Denver-Boulder-Greeley, CO Detroit-Ann Arbor-Flint, MI Houston-Galveston-Brazoria, TX Los Angeles-Riverside-Orange County, CA Miami-Fort Lauderdale, FL Milwaukee-Racine, WI New York-Northern New Jersey-Long Island, NY-NJ-CT-PA Philadelphia-Wilmington-Atlantic City, PA-NJ-DE-MD Portland-Salem, OR-WA Sacramento-Yolo, CA San Francisco-Oakland-San Jose, CA Seattle-Tacoma-Bremerton, WA Washington-Baltimore, DC-MD-VA-WV Source: Employee Benefit Research Institute estimates from the 2002 March Current Population Survey. 14 April 2003 EBRI Issue Brief

15 Figure 13 Percentage of Various Work Forces Who Worked for an Employer That Sponsored a Retirement Plan and the Percentage Who Participated in an Employment-Based Retirement Plan, All Workers (millions) Total Works for a sponsoring employer Participating in a plan (percentage) Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Works for a sponsoring employer Participating in a plan Wage and Salary Workers Ages (millions) Total Works for a sponsoring employer Participating in a plan (percentage) Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Works for a sponsoring employer Participating in a plan Private-Sector Wage and Salary Workers Ages (millions) Total Works for a sponsoring employer Participating in a plan (percentage) Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Works for a sponsoring employer Participating in a plan Public-Sector Wage and Salary Workers Ages (millions) Total Works for a sponsoring employer Participating in a plan (percentage) Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Works for a sponsoring employer Participating in a plan Full-Time, Full-Year Wage and Salary Workers Ages (millions) Total Works for a sponsoring employer Participating in a plan (percentage) Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Works for a sponsoring employer Participating in a plan Source: Employee Benefit Research Institute estimates from the March Current Population Surveys. April 2003 EBRI Issue Brief 15