Back to Blog List

Topics/Previous Posts

  • Stephen Crawford, Research Professor, George Washington University Institute of Public Policy
  • Comments (0)

A Point of View on Lifelong Learning and Upskilling

"The obvious solution is to improve and expand programs for “reskilling” and credentialing mid-career and older workers, but that is difficult. It is hard to know exactly what skills are needed, how to motivate workers to learn them, and how best to teach them. And it is very difficult to finance the training. Progress depends on each of the key stakeholders – employers, workers, institutions that provide education and training, and governments –doing its part, collaborating with the others, and sharing the financial burden."

Stephen Crawford

Research Professor, George Washington University Institute of Public Policy

Several studies show what most workers know: Aging is hazardous to their employment. According to a 2018 Urban Institute report, “[s]lightly more than one-half of full-time, full-year workers ages 51 to 54 with a long-term employer experienced an employer-related involuntary job separation after age 50 that led to a long-term unemployment spell or that reduced weekly earnings at least 50 percent for two or more years” (Johnson & Gosselin, 2018). Nearly two in five workers retire earlier than planned, which is tough on them and their families, and bad for the economy and tax revenue (Munnell & Waters, 2019).


Employers terminate older workers for a variety of reasons, but two major ones are the job’s changing skill requirements and a reluctance to invest in “reskilling” employees viewed as slow to learn new skills, especially technical ones, and likely to retire before the investment pays off (Heidkamp, 2012). On the employee side, a national survey of adult job seekers found 40 percent of respondents ages 50 and older felt that they lacked the skills for available jobs (Benz, et al, 2013).


This skill challenge has been an issue for American workers since the beginning of the industrial revolution but may be growing. A new wave of automation – robotics, artificial intelligence, etc. – is eliminating some jobs and creating others, just as mechanization and other innovations did in the past. More importantly, it is also changing, and usually increasing, the skills required to do existing jobs. Simultaneously, increased global competitiveness, rising health insurance costs, especially for older workers, and the decline of loyalty between employers and employees are motivating employers to dismiss older workers whose skills appear dated rather than upskill them or find them a less challenging job within the firm (Johnson & Gosselin, 2018; Eyster, et al, 2001; Waddoups, 2016; Burtless, 2017). i

Furthermore, the workforce is aging, with the share of workers 55 and over rising from 12.4 percent in 1998 to a projected 25.2 percent in 2028. The over-55 share is increasing not only because the population is aging, but because labor force participation by older cohorts is increasing in response to longer life expectancies and declines in the certainty of retirement income (Bureau of Labor Statistics, 2019). ii The result is that more mid-career and older workers than ever 11.5 million according to one recent projection face the risk of displacement by machines or younger workers whose technical skills are more up-to-date (Lund, et al, 2019).


The problem is one of skills and stereotypes about adaptability, but also of credentials, especially for workers seeking a new job. There has been tremendous growth in the number of jobs requiring or preferring a degree, certificate, certification, or license paired with considerable growth in the number of younger workers who have in-demand credentials (Cunningham, 2019). iii Although older  workers have many skills acquired through experience, including the increasingly important social-emotional ones, such skills are rarely documented by credentials (Cunningham & Villasenor, 2016). Moreover, the relative payoff to different credentials and fields of study is often opaque, and career advice tailored to older workers is rarely available.


The obvious solution is to improve and expand programs for “reskilling” and credentialing mid-career and older workers, but that is difficult. It is hard to know exactly what skills are needed, how to motivate workers to learn them, and how best to teach them. And it is very difficult to finance the training. Progress depends on each of the key stakeholders – employers, workers, institutions that provide education and training, and governments –doing its part, collaborating with the others, and sharing the financial burden.


Employers spend substantial sums on education and training, but mostly to improve recruitment and retention of entry-level workers in high-turnover jobs or on employees who already have a bachelor’s degree (Carnevale, et al, 2015). iv They are reluctant to invest in older workers for the reasons mentioned above, though research shows that many of the stereotypes are wrong and that under the right conditions, older workers can and will learn new skills and work past normal retirement age (Ng & Feldman, 2012; Picchio, 2015; Friedberg, 2003).


Employees understand that the world of work has changed. According to the Pew Research Center, 87 percent of them know it’s important to acquire new skills throughout their career to keep up with workplace changes (Pew, 2016). Yet, they are reluctant to invest in additional education and training for reasons of cost, time, anxiety about returning to the classroom, and uncertainty about providers and payoff. With relatively few years left in their careers, older workers in particular wonder whether it’s worth the expense, effort and uncertainty.


Successful reskilling requires effective education and training programs at community colleges and elsewhere, and accessible information about them. Some community colleges, notably the AACC’s “Plus 50+” ones and those engaged in AARP’s “Back to Work 50+” program are doing fine work. v Yet, many more need to offer short “stackable” courses, age-peer-group classes, online and competency-based instruction, “guided pathways” and strong coaching, all carefully aligned with the skill needs of employers in the region. Education and training providers should also improve their assessment techniques, offer credit for prior learning, and issue certificates for small chunks of progress.


None of this is likely, however, unless governments take the lead, not only by expanding their own programs but also by providing effective incentives to employers, workers, and education and training providers. The federal government offers a small, non-refundable Lifelong Learning [tax] Credit and is considering allowing the use of Pell grants for short-term training. Maine and Washington offer lifelong learning accounts, and many states provide incentives to employers for incumbent worker training. Still, governments could do much more. vi

The Aspen Institute recently proposed a worker-owned “Lifelong Learning and Training Account” that would allow tax-advantaged contributions by workers and employers as well as government matches (up to 50 percent) for low-income workers. It also proposed proactive retraining efforts for workers in at-risk jobs by providing their employers with a larger Worker Training Tax Credit (McKay, et al, 2019). The federal government could also offer wage insurance to encourage workers to shift to new jobs that may not pay as well but are more secure and offer better opportunities going forward.


The public sector should incentivize community colleges to make their occupational programs more attractive as suggested above, and to make greater use of asynchronous online instruction, virtual reality, and other technologies for increasing instructional convenience and effectiveness. More generally, the federal government should join other countries in investing much more in active labor market policies, especially retraining. Some critics question the effectiveness of publicly funded workforce training, and evaluations are mixed. But programs such as Project Quest in San Antonio have proved quite successful, and as MIT’s Paul Osterman emphasizes in a recent paper, the real challenge is taking best practices to scale (Osterman, 2019). vii

Technology-driven workplace change threatens many workers, especially older ones, but also brings new
opportunities. Emerging jobs may be less physically demanding, put a higher premium on social-emotional skills, and permit working from home. Emerging instructional technologies may make it easier to learn new skills on the job or at home rather than by attending classes. Key stakeholders, especially governments, will have to do more to change current expectations concerning training and credentialing for older workers and finance the needed reskilling and credentialing efforts.


Benz, J., Sedensky, M., Thompson, T., & Agiesta, J. (2013). Working Longer: Older Americans' Attitudes on Work and Retirement. Associated Press-NORC. Retrieved from Longer/AP-NORC Center_Working Longer Report-FINAL.pdf 


Bureau of Labor Statistics (BLS). 2019. Civilian labor force participation rate by age, sex, race, and ethnicity. Employment projections program. Retrieved from 


Burtless, G. (2017). Age Related Health Costs and Job Prospects of Older Workers. Retrieved from 


Carnevale, A. P., Strohl, J., & Gulish, A. (2015). College Is Just the Beginning: Employers’ Role in $1.1 Trillion Postsecondary Education and Training System. Georgetown University Center on Education and the Workforce. Retrieved from 


Cunningham, E. (2019, June 1). Professional certifications and occupational licenses: evidence from the Current Population Survey: Monthly Labor Review. U.S. Bureau of Labor Statistics.

Cunningham, W. V., & Villaseñor, P. (2016). Employer Voices, Employer Demands, and Implications for Public Skills Development Policy Connecting the Labor and Education Sectors. World Bank Research Observer, 31(1), 102–134. doi: 10.1093/wbro/lkv019    

Eyster, L., Johnson, R. W., & Toder, E. (2008). Current Strategies to Employ and Retain Older Workers. The Urban Institute. Retrieved from

Friedberg, L. (2003). The Impact of Technological Change on Older Workers: Evidence from Data on Computer Use. Industrial and Labor Relations Review, 56(3), 511-529. doi:10.2307/3590922

Heidkamp, M. (2012). Older Workers, Rising Skill Requirements and the Need for a Re-envisioning of the Public Workforce System. CAEL. Retrieved from

Johnson, R. W., & Gosselin, P. W. (2018). How Secure Is Employment at Older Ages? (pp. 1–20). Washington, DC: Urban Institute.

Lund, S., Manyika, J., Segel, L. H., Dua, A., Hancock, B., Rutherford, S., & Macon, B. (2019). The future of work in America: People and places, today and tomorrow. McKinsey Global Institute. Retrieved from 

McKay, C., Pollack, E., & Fitzpayne, A. (2019). Automation and a Changing Economy: Policies for Shared Prosperity. The Aspen Institute. Retrieved from 

Munnell, A. H., & Waters, A. N. (2019). Proposals to Keep Older People in the Labor Force. The Brookings Institution. Retrieved from

Ng, T. W. H., & Feldman, D. C. (2012). Evaluating Six Common Stereotypes About Older Workers with Meta‐Analytical Data. Personnel Psychology, 65(4), 821–858. doi: 10.1111/peps.12003 

Osterman, P. (2019). Employment and Training for Mature Adults: The Current System and Moving Forward. The Brookings Institution. Retrieved from

Pew Research Center. (2016). The State of American Jobs: How the shifting economic landscape is reshaping work and society and affecting the way people think about the skills and training they need to get ahead. Retrieved from

Picchio, M. (2015, January 1). Is training effective for older workers? IZA World of Labor: 121. doi: 10.15185/izawol.121

Waddoups, C. J. (2016). Did Employers in the United States Back away from Skills Training during the Early 2000s? Industrial and Labor Relations Review, 69(2), 405–434. doi: 10.1177/0019793915619904


i These other developments rather than ongoing skill obsolescence may explain the rising share of retirees who report that they were forced to retire. See Johnson and Gosselin, op. cit. for evidence on such reports. On the lower participation rates of older workers in employer-sponsored training programs, see L. Eyster, R. W. Johnson and E. Toder. On the decline in employees receiving training, see Jeffrey Waddoups. On employer health costs, a recent study found they run about $1,500 more a year for older workers than for younger ones; see Gary Burtless.

ii According to the Bureau of Labor Statistics, the share of workers age 55 and older in 1998 was 12.4 percent, but rose to 18.1 percent in 2008, 23.1 percent in 2018, and will likely be 25.2 percent in 2028. Labor force participation for those over 55 rose from 31.3 percent in 1998 to 40.0 percent in 2018.

iii There have been large increases in educational attainment in the U.S. over the past half-century, not only in degrees but also certificates and licenses. The percentage of adults over 25 with a master’s degree, 13 percent, is now the same as for those with a bachelor’s in 1973. The number of certificates has grown even faster. If we include certificates issued by industry associations and companies like Microsoft and Google, the number far exceed that of degrees. The percentage of the workforce that has a license has roughly quadrupled since 1960, but does not vary much by age.

iv 58 percent is spent on employees who already have a bachelor’s or more advanced degree, 25 percent on employees who have some college, and only 17 percent on employees who have only a high school diploma or less. Moreover, much of the spending that goes to employees who do not hold college degrees does not result in formal credentials or credit toward credential attainment.

v For information about the American Association of Community Colleges’ Plus 50 Initiative and evaluations of it, see: Plus 50 Programs in Practice, prepared by Learning for Action (ND). For information about the AARP’s Back to Work 50+ program and its subsequent version for older women, see Betesh, et al, 2017. There are providers other than community colleges. For a good discussion, including of providers other than community colleges, see Paul Osterman, 2019.

vi The current Chair’s Initiative of the National Governors Association focuses on what governors can do to “help midcareer workers succeed in the workforce through increased access to continuous learning opportunities, improved career transitions and holistic support.” See A Governor’s Action Guide to Achieving Good Jobs for All Americans, 2019. See also Robert Sheets and Stephen Crawford (2016). Several states are developing definitions of quality for the confusing array of non-degree credentials (Duke-Benfield, et al, 2019).

vii Many existing workforce development programs have been evaluated as effective using the gold-standard Random Control Trials, but that what is effective in one time and place will not necessarily scale, because contexts are important. Governments can help by providing adaptation guidelines; see Alexander Ruder, 2019. 

Leave a Comment

Only comments approved by post author will be displayed

Back to Blog List