Workers without employer-provided retirement savings plans risk negative economic consequences in their retirement years. They may enter retirement without adequate savings or adequate Social Security benefits. They may face impoverished retirements or may even be unable to retire. They may also be more likely to turn to government programs such as Medicaid and Supplemental Security Income, straining state and federal budgets. What is more, such workers may lack a buffer, in the form of retirement savings, during a financial crisis either before or in retirement.
Lack of access to a workplace plan is the most significant barrier to retirement savings facing nontraditional workers, also known as contingent, gig, alternative, or independent workers. Nontraditional workers are different from workers in a traditional employer-employee relationship. They don’t readily fall into simple categories, they may lack job security, their incomes may be volatile, and they generally do not enjoy employer-provided benefits such as health insurance or a retirement plan.
To help inform policymakers considering retirement savings options for nontraditional workers, in 2020 The Pew Charitable Trusts surveyed 1,000 workers who said they worked at nontraditional jobs. The survey examines their access to workplace retirement plans, the barriers they face in saving for retirement, and the types of programs— existing or new—that would work best for them. This is the third report using Pew’s survey of nontraditional workers to examine their retirement security. A first report examined nontraditional work during the pandemic,1 and a second looked at nontraditional workers’ savings balances in workplace retirement plans and IRAs.2 Subsequent reports will explore survey results to look at barriers to saving for retirement, couples’ coordination of retirement savings, nontraditional workers’ understanding of financial concepts and skills, and which types of retirement savings programs appeal to different groups of nontraditional workers.
Multiple, innovative savings solutions will likely be needed to help very different types of nontraditional workers save for retirement. State-facilitated auto-IRA programs, which enroll private sector workers in a savings plan when they do not have a workplace plan, are one potential solution that could be adapted to reach nontraditional workers. Other possible approaches could include encouraging savings through the tax system or through a financial institution or financial technology (“fintech”) apps that help automate retirement savings transactions.
This report adopts a definition of nontraditional work that is based on arrangements between workers and the individual or company that pays them. The survey recruited respondents who reported the following types of work relationships: online or platform work, temporary help agency work, freelancers and consultants, contract company work, sole proprietors, on-call work, day labor, seasonal work, and other nontraditional work. Pew conducted focus groups before fielding the survey in order to refine the types of nontraditional work into categories that were meaningful to focus group participants. “Traditional workers” in this report earn a wage or salary.
These types of arrangements are sometimes called contingent, gig, alternative, independent, or nonstandard work arrangements.
Nontraditional work comes in many forms—freelancing, sole proprietorship, online or platform work, oncall work, contract work, temporary help agency work, partnerships, and more. Labor experts agree that nontraditional work’s common denominator is a lack of job security, but consensus often ends there. Varying definitions of nontraditional work, and different datasets, give rise to a wide range of estimates, from 3.8% of America’s workforce to as much as 40.4%.6
Pew recruited survey participants from a range of nontraditional work arrangements, and these work categories drive our definition of “nontraditional work.” 7 (See “Who Are Nontraditional Workers?”) Because nontraditional jobs involve a variety of work situations, Pew’s survey uses work characteristics to develop three types of nontraditional workers and views workers’ retirement security through these work types. The three main types of nontraditional work in the analysis are (1) a single nontraditional job, (2) a mix of traditional and nontraditional work, and (3) multiple nontraditional jobs. Workers in traditional employee-employer relationships were included in Pew’s survey if they also performed nontraditional work. Many nontraditional workers had multiple jobs, and the survey asked detailed information for the three most important jobs. Pew did not attempt to measure the nontraditional workforce; instead, our goal was to understand nontraditional workers’ experience with retirement savings.
Because retirement plan access and interest vary considerably by whether the worker has a traditional job and whether they have one or multiple nontraditional jobs, Pew divided workers into three categories. (See “Three Worker Categories.”) As shown in Figure 1, 37.6% of the nontraditional workers had a single nontraditional job (“solo nontraditional”). The other workers held multiple jobs: 40% of nontraditional workers had a mix of nontraditional and traditional jobs (“mixed traditional-nontraditional”), and the remaining quarter (22.4%) had two or three nontraditional jobs but no traditional job (“multiple nontraditional”).
Some workers rely on nontraditional work more than others. To better understand how workers fit nontraditional work into their total earnings, Pew divided the sample into three categories. Solo nontraditional workers hold a single, nontraditional job. Mixed traditional-nontraditional workers hold multiple jobs that include a mix of traditional and nontraditional jobs, where traditional jobs were performed on a full-time or part-time basis. The final group holds multiple nontraditional jobs but no traditional job.
The various forms of nontraditional work have different demographic attributes, underscoring why almost any single impression or stereotype of nontraditional work is wrong. Over half (59.7%) of all nontraditional workers were White, 12.3% were Black, and 19.3% were Hispanic (Table 1).8 However, demographic characteristics within different worker categories show somewhat different shares. Relative to their shares of the survey population, solo nontraditional workers were more likely to be White, to be married or partnered, to have very high (over $200,000) household incomes, to be over age 50, and to report having done this kind of work for five or more years. Workers with mixed nontraditional-traditional jobs were more likely to be Black, to be single, to have completed some college, and to report doing this work for less than five years. Workers with multiple nontraditional jobs were more likely to be male, to be non-White, to have only a high school diploma or less, and to report doing this work for five years or more. The great majority (82.3%) of nontraditional workers spent 20 or more hours per week at their primary job (the job that provided the most income, whether traditional or nontraditional).
Table 1
| Total | Solo nontraditional job | Mixed nontraditional/traditional jobs | Multiple nontraditional jobs | |
|---|---|---|---|---|
| Gender | ||||
| Male | 54.0% | 50.0% | 53.2% | 61.9% |
| Female | 46.0% | 50.0% | 46.8% | 38.1% |
| Race/ethnicity | ||||
| White, non-Hispanic | 59.7% | 67.4% | 58.5% | 48.7% |
| Black, non-Hispanic | 12.3% | 9.4% | 14.7% | 13.1% |
| Hispanic | 19.3% | 16.2% | 16.9% | 28.7% |
| Other | 8.7% | 7.0% | 9.9% | 9.5% |
| Marital status | ||||
| Coupled (married or partnered) | 56.0% | 62.7% | 51.0% | 53.4% |
| Single (separated, divorced, widowed, or never married) | 44.0% | 37.3% | 49.0% | 46.6% |
| Education | ||||
| High school or less | 35.9% | 39.9% | 27.9% | 43.4% |
| Some college | 25.0% | 20.5% | 31.6% | 20.9% |
| Bachelor’s degree | 22.9% | 20.1% | 24.0% | 25.6% |
| Graduate degree (M.A., M.S., Ph.D., etc.) | 16.3% | 19.5% | 16.6% | 10.1% |
| Age | ||||
| Age 50 and over | 36.1% | 48.1% | 25.2% | 35.2% |
| Age 60 and over | 18.8% | 28.1% | 8.9% | 20.8% |
| Household income | ||||
| Less than $5,000 | 3.1% | 3.4% | 3.0% | 3.1% |
| $5,000 to $19,999 | 12.5% | 11.9% | 10.0% | 18.0% |
| $20,000 to $49,999 | 26.8% | 26.6% | 27.4% | 26.2% |
| $50,000 to $99,999 | 34.7% | 35.3% | 35.5% | 32.1% |
| $100,000 to $149,999 | 14.5% | 14.5% | 14.5% | 14.1% |
| $150,000 to $199,999 | 4.7% | 3.0% | 6.4% | 4.5% |
| $200,000 and higher | 3.7% | 5.3% | 3.1% | 2.1% |
| Number of years doing nontraditional work | ||||
| Fewer than five years | 54.8% | 47.5% | 68.7% | 42.0% |
| Five or more years | 45.2% | 52.5% | 31.3% | 58.0% |
| Primary job: hours worked per week | ||||
| 1-19 hours | 17.7% | 17.3% | 13.4% | 26.1% |
| 20 or more hours | 82.3% | 82.7% | 86.6% | 73.9% |
Notes: Sample size=1,026 for all demographic and income categories. Sample size=997 for tenure in nontraditional work. “Other” race/ethnicity includes Native Americans, Asians and Pacific Islanders, and workers who reported two or more categories of race/ethnicity. Because of rounding, some categories do not add up to 100%. © 2021 The Pew Charitable Trusts
Nontraditional workers described in detail the jobs that brought them the most income, up to a maximum of three jobs. If the respondent had a single job, that job was in nontraditional work as a condition of participating in the survey (Table 2). More than half of respondents with two or more jobs had a traditional primary job, defined as the job that brought them the most income: 53.6% of workers with two jobs had a traditional primary job, as did 55.6% of workers with three jobs, highlighting the role of online and gig work in providing supplemental income.
The variety of jobs performed by nontraditional workers is demonstrated in Table 2, not just in the list of jobs but in the fact that 18.6% cited “other” contingent work as their primary job. Roughly one-third (34.9%) of nontraditional workers with a single job were sole proprietors, 16.1% did contract company work, and 10.9% were freelancers. Moreover, 12.3% of those with two jobs, and 14.3% of those with three jobs, reported that a sole proprietorship was their primary source of income. Freelancing was a key source of additional income for many with two or more jobs, again underscoring the importance of supplemental income for nontraditional workers. In contrast, online or gig work (e.g., Uber, Lyft, Etsy, eBay, Observa, TaskRabbit, or Upwork) was the sole source of income for just 3.7% of nontraditional workers, and it was the primary source of income for only 5.4% of those with two jobs, and for 5.2% of those with three jobs. But 15.1% of workers with two jobs said online or gig work was their second-most important source of income, and 21.7% of workers with three jobs said online or gig work was their second- or third-most important source of income.
Looking across all nontraditional workers with one or multiple jobs, almost a third (32%) said their primary source of income was from traditional work (not shown). Put another way, a third of all nontraditional workers received the greatest share of their income from traditional employment and moonlighted with nontraditional work.
Table 2
| Respondent has two jobs | Respondent has three jobs | ||||||
|---|---|---|---|---|---|---|---|
| Respondent has one job | Primary source of income | Second-most important source of income | Primary source of income | Second-most important source of income | Third-most important source of income | All nontraditional workers: primary source of income | |
| Traditional job | 0.0% | 53.6% | 8.8% | 55.6% | 26.6% | 10.0% | 32.0% |
| Online or gig worker | 3.7% | 5.4% | 15.1% | 5.2% | 9.9% | 11.8% | 4.9% |
| Temporary help agency worker | 4.7% | 4.0% | 5.7% | 3.7% | 2.8% | 5.0% | 4.0% |
| Freelancer | 10.9% | 4.3% | 20.5% | 4.3% | 17.0% | 19.2% | 7.8% |
| Contract company worker | 16.1% | 7.6% | 7.1% | 6.7% | 8.1% | 3.4% | 10.5% |
| Sole proprietor | 34.9% | 12.3% | 14.5% | 14.3% | 10.7% | 10.4% | 21.6% |
| On-call worker | 2.7% | 2.6% | 6.9% | 0.7% | 4.3% | 7.0% | 2.1% |
| Day laborer | 2.6% | 1.8% | 1.8% | 2.5% | 3.8% | 8.0% | 2.6% |
| Seasonal worker | 5.9% | 1.3% | 7.0% | 3.9% | 8.5% | 15.7% | 3.4% |
| Other | 18.6% | 7.1% | 12.6% | 3.1% | 8.3% | 9.5% | 11.1% |
| Total | 100.1% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% | 100.0% |
Notes: Respondents with multiple jobs were asked to identify their primary source of income (“primary job”), and their second- and third-most (if applicable) important sources of income. Online or gig work includes labor and capital platforms such as Uber, Lyft, Etsy, eBay, Observa, TaskRabbit, or Upwork. © 2021 The Pew Charitable Trusts
For nontraditional workers, lack of access to a workplace plan is the most significant barrier to retirement savings they face. When a worker has an employer-provided DC plan, the automatic nature of payroll contributions means that he or she doesn’t have to remember to save and contributions are made before the worker can rethink them. Additionally, employer-provided DC plans often use behavioral nudges, such as auto-enrollment, default contribution rates, and sometimes even auto-escalation of contribution rates to promote worker savings.
Roughly half (46.3%) of all nontraditional workers had a job or an employer during the previous year that offered a DB, DC, or other type of retirement plan such as an annuity. Access to a plan was strongly tied to having at least one traditional job: Only 29.8% of nontraditional workers with no traditional job had access to a workplace retirement plan. By comparison, 69.7% of nontraditional workers who also had one or more traditional jobs had a workplace plan such as a DB, DC, or other plan.9
Approximately one-third (33.9%) of all nontraditional workers had a workplace that offered a DC plan over the past year, whether or not they were eligible for the plan or participated in it.10 Those whose workplaces offered a DC plan were more likely to be male, to be Hispanic or “other” race or ethnicity (including Asians and workers with two or more ethnicities), to have a college degree, to have higher incomes, to have been doing nontraditional work for fewer than five years, and to work 20 or more hours a week.11
Table 3
| Worker characteristics | Access to DC plan (%) |
|---|---|
| All | 33.9% |
| Gender | |
| Male | 36.4% |
| Female | 31.0% |
| Race/ethnicity | |
| White, non-Hispanic | 31.3% |
| Black, non-Hispanic | 30.6% |
| Other | 55.4% |
| Marital status | |
| Coupled (married or partnered) | 34.5% |
| Single (separated, divorced, widowed, or never married) | 33.1% |
| Education | |
| High school or less | 24.0% |
| Some college | 38.1% |
| Bachelor’s degree | 42.3% |
| Graduate degree (M.A., M.S., Ph.D., etc.) | 37.5% |
| Age | |
| Age 50 and over | 29.2% |
| Age 60 and over | 20.2% |
| Household income | |
| Less than $5,000 | 14.9% |
| $5,000 to $19,999 | 20.0% |
| $20,000 to $49,999 | 31.4% |
| $50,000 to $99,999 | 37.2% |
| $100,000 to $149,999 | 39.7% |
| $150,000 to $199,999 | 45.8% |
| $200,000 and higher | 46.3% |
| Number of years doing nontraditional work | |
| Fewer than five years | 38.8% |
| Five or more years | 27.9% |
| Primary job: hours worked per week | |
| 1-19 hours | 17.9% |
| 20 or more hours | 37.5% |
Notes: “Other” is NORC’s category and includes Asians, Pacific Islanders, American Indians, and workers with two or more races or ethnicities. Sample size=1,026 for all demographic variables. Sample size=997 for the number of years in nontraditional work, and 1,006 for hours worked per week.
© 2021 The Pew Charitable Trusts
When a workplace offered a DC plan, a high proportion (83.4%) of nontraditional workers were eligible for it. Time spent in a job affects eligibility for workplace benefits because employers often impose a waiting or service period before a worker can join the plan. Workers who had been doing nontraditional work for five or more years were more likely (88.1%) to be eligible for a workplace plan than those with fewer than five years of work (79.8%). When nontraditional workers were eligible for a workplace plan, 77.5% took the opportunity to participate in it.
Putting all the numbers—access, eligibility, and takeup—together in terms of the overall workforce, only 21.9% of all nontraditional workers participated in a workplace DC plan (Figure 2).12 Worker access to and participation in a workplace DC plan were strongly tied to whether the worker had a traditional job as one source of income. More workers with mixed traditional-nontraditional jobs participated in a workplace DC plan (39.3%) than solo nontraditional workers (12.3%) or workers with multiple nontraditional jobs but no traditional job (7.3%).
The finding that over three-quarters of eligible nontraditional workers took up a workplace plan—when offered and eligible for one—indicates strong demand among nontraditional workers for such plans. This result is consistent with a second survey question that asked nontraditional workers which workplace benefits they wanted, with multiple choices allowed: Two-thirds (66.3%) of nontraditional workers said they wanted retirement benefits. In fact, retirement benefits took second place on workers’ benefits wish lists after health benefits, which were desired by 76.6% of nontraditional workers.
Statistical analysis in the form of logistic regression was used to supplement the cross-tabulations used in this issue brief. The dependent variable explained in the regression analysis is access to a workplace DC plan. The analysis allows for assessing the relationship between access to a workplace DC plan and each individual characteristic, such as job mix, education, or race and ethnicity, while holding other characteristics constant.
A worker’s job mix (whether he or she had single or multiple nontraditional jobs, or a traditional job in addition to a nontraditional job or jobs) was a significant effect: Relative to those with a mix of traditional-nontraditional work, workers in both solo nontraditional and multiple nontraditional jobs were much less likely to have access to a DC savings plan. In addition, higher educational attainment and Hispanic ethnicity13 were associated with greater access to a DC plan, after controlling for other demographic and work factors. For the modeling results, see Appendix B.
Nontraditional workers appear to be very interested in gaining access to retirement savings plans, and low participation by this group in workplace retirement savings is more often due to a lack of access than to a lack of demand. Less than half (46.3%) of nontraditional workers had a workplace that offered any type of retirement plan (DB, DC, or other). When a DC plan was offered to these workers, most were eligible for it, although 16.6% were not. Takeup rates were very high, however, and 77.5% of nontraditional workers chose to participate in a workplace DC plan when one was offered and they were eligible. Along similar lines, when asked to indicate which workplace benefit they wanted, with multiple choices allowed, two-thirds (66.3%) of nontraditional workers said they wanted retirement benefits. In fact, retirement benefits took second place on workers’ benefits wish lists after health benefits, which were desired by 76.6% of nontraditional workers.
Access to a retirement plan was associated with work that mixed traditional employer-employee relationships with nontraditional work (“mixed traditional-nontraditional”), which makes sense given that plans offered as part of traditional employment relationships are the chief vehicle for retirement savings in the U.S. The lowest rates of access were among those workers who cobbled together more than one nontraditional job (“multiple nontraditional jobs”).
Solutions are needed that provide retirement savings opportunities directly and efficiently to all workers. If given the opportunity, many workers will save. The question, then, is what might facilitate savings by nontraditional workers who have a variety of work situations? Future work from this Pew survey will explore in more depth the barriers that nontraditional workers face to saving for retirement, such as other demands on income or not understanding how to save for retirement; the potential for coordination with a spouse or partner who might have access to a workplace retirement plan; and alternative approaches to retirement saving. Such approaches may be through policy initiatives such as state-facilitated auto-IRA programs that enroll private sector workers in a savings plan when they do not have a workplace plan, or by encouraging savings through the tax system, a financial institution, or a fintech app. It’s very possible that more than one solution may be necessary to address the varied needs of this diverse segment of the labor force.
The following table highlights the relationships between the job groups used in this report (solo nontraditional job, mixed traditional-nontraditional jobs, and multiple nontraditional jobs) and the various types of jobs that survey participants held as their primary source of income (a traditional job, online or platform work, temporary help agency work, etc.).
Thus Table A1 shows the primary job, defined as the job that is the worker’s most important source of income, for all nontraditional workers in the study. Among solo nontraditional workers (first column), the primary job is likely to be in freelancing, contract work, a sole proprietorship, or “other” types of nontraditional work. Workers with mixed nontraditional-traditional work (second column) are very likely to have traditional work as their primary job. And finally, when the worker has multiple nontraditional jobs, the primary job is predominantly contract work, a sole proprietorship, or “other” nontraditional work, with a healthy sprinkling of online/gig work and freelancing.
Table A1
| Solo nontraditional job | Mixed nontraditional/traditional jobs | Multiple nontraditional jobs | Total | |
|---|---|---|---|---|
| Traditional job | 0.0% | 80.0% | 0.0% | 32.0% |
| Online or gig worker | 3.8% | 2.4% | 11.2% | 4.9% |
| Temporary help agency worker | 4.7% | 0.9% | 8.3% | 4.0% |
| Freelancer | 10.9% | 3.8% | 9.7% | 7.8% |
| Contractor | 16.1% | 3.3% | 14.1% | 10.5% |
| Sole proprietor | 34.9% | 5.4% | 28.3% | 21.6% |
| On-call worker | 2.7% | 0.4% | 4.2% | 2.1% |
| Day laborer | 2.6% | 1.0% | 5.6% | 2.6% |
| Seasonal worker | 5.9% | 0.7% | 4.0% | 3.4% |
| Other | 18.6% | 2.1% | 14.6% | 11.1% |
| Total | 100.2% | 100.0% | 100.0% | 100.0% |
Note: Sample size=1,026.
© 2021 The Pew Charitable Trusts
The variable being explained in the regression analysis (dependent variable) is access to a workplace DC plan (Table B1). Independent variables, which help to explain the access to a workplace plan, include the demographic characteristics of gender, race and ethnicity, educational attainment, and income. Two work characteristics (whether the worker has a traditional job or single or multiple nontraditional jobs, and how long the worker has been doing contingent work) also serve as independent variables.
Odds ratios are interpreted in terms of probability, with odds ratios greater than 1 indicating higher odds than the reference case, and odds ratios less than 1 indicating lower odds than the reference case. Thus, if a worker received a college or graduate degree, his or her odds of having access to a workplace DC plan were 1.7 times greater than a worker who had a high school education or less, holding all other variables constant. If a worker had a single nontraditional job, the odds of having a workplace DC plan decreased by a factor of 0.226 relative to a worker with a mix of traditional and nontraditional jobs, holding all other variables constant. Note that these interpretations contain no information about the magnitude of the implied change in probability.
Table B1
| Worker group | Odds ratio | Standard error |
|---|---|---|
| Mix of traditional and nontraditional jobs | (reference) | |
| Single nontraditional job | 0.226 | 0.049 |
| Multiple nontraditional jobs | 0.137 | 0.042 |
| Gender | ||
| Male | (reference) | |
| Female | 0.853 | 0.163 |
| Race/ethnicity | ||
| White, non-Hispanic | (reference) | |
| Black, non-Hispanic | 1.184 | 0.329 |
| Hispanic | 1.837 | 0.477 |
| Other | 3.091 | 1.328 |
| Education | ||
| High school or less | (reference) | |
| Some college | 1.403 | 0.362 |
| College or graduate degree | 1.707 | 0.461 |
| Income | ||
| Less than $5,000 | (reference) | |
| $5,000 to $19,999 | 1.494 | 1.108 |
| $20,000 to $49,999 | 2.479 | 1.715 |
| $50,000 to $99,999 | 3.128 | 2.142 |
| $100,000 to $149,999 | 3.299 | 2.312 |
| $150,000 to $199,999 | 4.069 | 3.182 |
| $200,000 or more | 5.041 | 4.202 |
| How long doing contingent work | ||
| Fewer than five years | (reference) | |
| Five or more years | 0.823 | 0.159 |
| Constant | 0.3038484 | 0.20939 |
© 2021 The Pew Charitable Trusts
The Pew Charitable Trusts hired NORC at the University of Chicago to survey nontraditional workers to better understand their access to retirement savings plans through work or outside of work. The sample was drawn from NORC’s nationally representative AmeriSpeak Panel. The survey was fielded online and by telephone, in English and Spanish, from June 4, 2020, to July 1, 2020. The survey obtained 1,026 qualified interviews from individuals age 18 and over who worked in nontraditional jobs (also referred to as contingent, gig, nonstandard, or independent jobs). Quotas were used to ensure enough completed surveys for three subtargets: workers with a single nontraditional job, workers with a mix of traditional and nontraditional jobs, and workers with multiple nontraditional jobs (but no traditional job). Survey results were weighted to reflect the selection probabilities of the panel members as well as weighting adjustments to ensure that the weighted panel represents the U.S. household population. The study sample supports proportion estimates with a margin of error no greater than 4.26 percentage points.
For more information on methods, see the survey methodology statement14 and topline results.15
The survey referenced in this brief was funded in part by The Pew Charitable Trusts with additional support from the FINRA Investor Education Foundation.