Laid off after 20 years in one industry: what the reemployment data shows

What displaced means in the official data, and how many people it covers

The Bureau of Labor Statistics runs a Displaced Workers survey every two years as a January supplement to the Current Population Survey. It defines displaced workers as people aged 20 and over who lost or left a job because the plant or company closed or moved, there was insufficient work, or their position or shift was abolished. That definition is deliberately narrow. It excludes people who quit, were fired for cause, or were let go from a job they had held briefly. The most recent release, published on 27 August 2026, covers displacements between January 2023 and December 2025. It counted 3.3 million long-tenured displaced workers, meaning three or more years with the employer, plus 4.1 million short-tenured, for 7.4 million in total, up from 6.3 million in the previous survey. Among the long-tenured group, 44.4 percent lost their job because the position or shift was abolished, 32.6 percent because the plant or company closed or moved, and 22.9 percent because of insufficient work. Nineteen percent came out of manufacturing, 16 percent out of professional and business services, and 10 percent out of retail trade. If your role was abolished rather than your employer closing, you are in the largest single category, and the survey treats you as a standard case rather than an unusual one.

The reemployment rate by age is the number to plan around

At the January 2026 survey date, 66.1 percent of the 3.3 million long-tenured displaced workers were reemployed, 18.3 percent were unemployed, and 15.7 percent had left the labour force. Those totals hide a steep age gradient. Reemployment was 72.9 percent for those aged 25 to 54, 57.3 percent for those aged 55 to 64, and 38.6 percent for those aged 65 and over. Men and women in the group were close, at 68.0 percent and 64.0 percent. Reemployment rates by major occupational group were tight, between 65.2 and 69.2 percent, which means the occupation you came from tells you less than your age does. Two cautions on reading this. The survey is a snapshot taken in January 2026, so someone displaced in December 2025 had weeks to look and someone displaced in early 2023 had three years. It is not a time-to-hire measure. And the not-in-labour-force share for older workers includes people who retired, willingly or otherwise, so the 57.3 percent figure for ages 55 to 64 is not a straightforward failure rate. What it does establish is that a substantial minority of people in your position were still not in work when the survey reached them, so a long search is the normal experience in the data rather than evidence that something is wrong with your applications. It does not establish how long that search runs, and no figure on this page does.

The pay cut is the part nobody warns you about

Of the 1.9 million long-tenured workers who lost full-time wage and salary jobs in this period and were reemployed by January 2026, 1.6 million had full-time jobs again. Among those who reported earnings on the job they lost, about 49 percent were earning as much or more than before. In the January 2024 survey that figure was about 62 percent, so the picture got worse rather than better. The published table gives the distribution. Of roughly 1.34 million reemployed full-time workers who reported earnings on both jobs, 369,000 were earning 20 percent or more below their lost job, 317,000 were earning less but within 20 percent, 354,000 were earning equal or above but within 20 percent, and 302,000 were earning 20 percent or more above. So about 27 percent took a cut of at least a fifth, and about 22 percent came out at least a fifth ahead. Both tails are real. Plan for the middle of that distribution rather than either end, and if you are running severance maths, model a replacement salary below your old one rather than at it. That single assumption changes how long your runway actually is.

Why the loss is durable, according to the long-run studies

The BLS snapshot shows the first year or two. Administrative earnings records show what happens over decades, and the picture is worse. Steven Davis and Till von Wachter, using Social Security records from 1974 to 2008, studied men aged 50 or younger with three or more years of tenure who lost jobs in mass-layoff events between 1980 and 2005. For that group, displacement reduced the present value of earnings by an estimated 77,557 dollars in 2000 dollars over the following 20 years, discounted at 5 percent a year, equivalent to 1.7 years of predisplacement earnings. The loss depended heavily on when you were displaced: 1.4 years of earnings if national unemployment was below 6 percent at the time, and 2.8 years if it was above 8 percent. The authors state that losses are larger for men with greater job tenure, which is the sentence that applies most directly to someone with twenty years in one place. They also summarise work by von Wachter, Song and Manchester finding that early-1980s displacements produced average annual earnings losses of more than 30 percent against a control group, and that even 20 years later displaced workers earned 15 to 20 percent less than the control group implied. These estimates cover men and are drawn from earlier decades, so read them as the shape of the problem rather than a forecast of your own number.

Advance notice did not change the outcome, which tells you something

About 45 percent of long-tenured displaced workers in the 2023 to 2025 period received written advance notice that their jobs would end. That rose to about 58 percent among workers whose plant or company closed or moved, and fell to about 42 percent where the position or shift was abolished and about 32 percent where the cause was insufficient work. Here is the finding worth sitting with: BLS reports that for each of these displacement groups, reemployment rates were not statistically different between those who received written advance notice and those who did not. Months of warning did not translate into a better landing. That result is not an argument for ignoring a notice period. It is an argument against the assumption that the notice period you were given was the resource that mattered. If advance warning by itself moved the number, it would show up here, and it does not. What that suggests is that the constraint sits somewhere else: in whether local employers are hiring for what you can already do, and whether you can reach a different set of adverts. Those are the two things you can act on, and neither of them is a function of how much notice you got.

What retraining measurably does, and what it does not

The closest thing to an answer comes from a federal retraining programme that has actually been evaluated. Trade Adjustment Assistance paid for retraining for workers who lost jobs to import competition, so it covers a narrower group than displaced workers generally, but it is the evidence that exists. Schochet, D'Amico, Berk, Dolfin and Wozny compared 2,054 TAA participants with 1,796 matched unemployment insurance claimants laid off in the same period and areas, average age 49. The training worked as training: the share who completed an education or training programme was 31 percentage points higher among participants, and the share who earned a certificate or degree was 30 percentage points higher. The earnings did not follow. Average annual earnings of TAA participants were significantly lower than the comparison group in all four years after the claim. The comparison group was matched rather than randomly assigned, and the Department of Labor's evidence review rates the causal evidence moderate on that account. The gap narrowed from 12,674 dollars in year one to 3,273 dollars in year four, but it never closed. Part of that is mechanical, since time spent in a classroom is time not spent earning, and the trend was toward convergence. But four years on, the credentialed group was still behind. The US Department of Labor also commissioned a randomised evaluation of the WIA Adult and Dislocated Worker programmes; its public landing page describes the design without stating the training impact, so I am not quoting a number from it. Treat any claim that retraining restores your old salary as unsupported by the evaluations that exist.

Where a certification fits after twenty years in one industry

Given all of that, a certification is worth buying in one specific circumstance: when it is the entry ticket to a set of job adverts you currently cannot answer, and those adverts exist within your commuting distance in numbers. It is not worth buying to signal seriousness, to fill time, or to make a resume look current. Your twenty years already carry more weight with a hiring manager than any entry-level credential, and the audit evidence on displaced applicants concerns how employers read the gap, which an exam does not address. The useful test is mechanical. Open the last thirty job adverts you would accept within an hour of home. Count how many name a specific credential as required rather than preferred. If the answer is under five, the credential is not what is stopping you and the money is better kept as runway. If the answer is fifteen or more, and they name the same one, you have found a gate rather than a preference, and paying to pass through it is a defensible use of a few hundred dollars. That is the whole decision. Everything else is a training provider's marketing.

A decision rule for the next 90 days

Order the moves by cost. First, work out the salary you can actually accept, using a replacement figure below your old one, because about half of reemployed full-time workers in the BLS data earn less than they did. The median annual wage for all US workers was 50,980 dollars in May 2025, which is a useful floor to sanity-check against. Second, do the thirty-advert count described above before you spend anything on training. Third, if a credential does gate the roles you want, check its shelf life before you enrol. CompTIA's Security+ exam SY0-701 launched in November 2023, runs 90 minutes with a maximum of 90 questions and a passing score of 750 on a 100 to 900 scale, and CompTIA states that versions usually retire about three years after launch, estimated 2026 for this one. Buying a course built around a version that is about to be replaced is a common and avoidable waste. Fourth, if you are choosing between a fast credential and a slower licensed occupation, price the slower one honestly, because the TAA evaluation shows people who completed real training were still behind on earnings four years later. None of this makes the loss disappear. It keeps you from paying twice for it.

What this does not fix

This page will not claim that retraining recovers the earnings you lost. The measured evidence points the other way: in the federal evaluation of Trade Adjustment Assistance, participants completed more training and earned more certificates than matched comparison workers, and were still earning less than them four years later. That evaluation is a matched-comparison design rather than a randomised one, so read it as strong evidence rather than proof. A certification changes which job adverts you are eligible to answer. It does not change the wage those adverts pay, and it does not undo the displacement penalty documented in the administrative earnings records.

Where an exam fits

• CompTIA A+, if help desk adverts near you name it as required and you accept that it is a bridge role paying below what you earned
• CompTIA Security+, if fifteen or more local adverts name Security+ by title, and only after checking which exam version is current
• CAPM (Certified Associate in Project Management), if you coordinated delivery for years without the title, the local adverts ask for CAPM, and you accept that it is an entry-level credential your twenty years already outrank
• CPC (Certified Professional Coder), if you are moving into healthcare administration and the coding job adverts in your area name the CPC specifically
• PTCB (Pharmacy Technician Certification), if your state regulates pharmacy technicians and you have checked what your own state board requires before paying
• Lean Six Sigma Green Belt (IASSC ICGB), if you came out of manufacturing or operations and the process improvement adverts near you name a belt level as required, since nothing on this page shows the credential raising pay on its own

Common questions

Sources

1. 01Number of long-tenured workers displaced from jobs held three or more years, January 2023 to December 2025 3.3 million long-tenured, up by 746,000 on the prior survey period, plus 4.1 million short-tenured, for 7.4 million total, up from 6.3 million in the 2021-23 survey
   US Bureau of Labor Statistics, Worker Displacement: 2023-2025, USDL-26-1423, released 27 August 2026 checked 2026-08-28
2. 02Labour force status of long-tenured displaced workers at the January 2026 survey date, overall and by age 66.1 percent reemployed, 18.3 percent unemployed and 15.7 percent not in the labour force; reemployment was 72.9 percent aged 25 to 54, 57.3 percent aged 55 to 64 and 38.6 percent aged 65 and over
   US Bureau of Labor Statistics, Worker Displacement: 2023-2025 checked 2026-08-28
3. 03Share of reemployed full-time long-tenured displaced workers earning as much or more than on the job they lost About 49 percent in January 2026, down from about 62 percent in the January 2024 survey
   US Bureau of Labor Statistics, Worker Displacement: 2023-2025 checked 2026-08-28
4. 04Distribution of new earnings relative to the lost job among reemployed full-time long-tenured displaced workers who reported earnings on both 369,000 earning 20 percent or more below, 317,000 below but within 20 percent, 354,000 equal or above but within 20 percent, and 302,000 20 percent or more above, out of about 1.34 million
   US Bureau of Labor Statistics, Displaced Workers Table 7, January 2026 checked 2026-08-28
5. 05Reason for job loss among long-tenured displaced workers, 2023 to 2025 44.4 percent position or shift abolished, 32.6 percent plant or company closed or moved, 22.9 percent insufficient work
   US Bureau of Labor Statistics, Worker Displacement: 2023-2025 checked 2026-08-28
6. 06Effect of written advance notice of termination on later reemployment About 45 percent of long-tenured displaced workers received written advance notice, rising to about 58 percent where the plant or company closed or moved, and falling to about 42 percent where the position or shift was abolished and about 32 percent where the cause was insufficient work. For each of these displacement groups, reemployment rates were not statistically different between those who received notice and those who did not
   US Bureau of Labor Statistics, Worker Displacement: 2023-2025 checked 2026-08-28
7. 07Median employee tenure by age, January 2024 9.6 years for workers aged 55 to 64, more than three times the 2.7 years for workers aged 25 to 34; 52 percent of workers aged 60 to 64 had 10 or more years with their current employer; the all-worker median was 3.9 years
   US Bureau of Labor Statistics, Employee Tenure in 2024, USDL-24-1971 checked 2026-08-28
8. 08Present value of earnings lost by long-tenured men displaced in mass-layoff events, from Social Security records covering 1974 to 2008 For men aged 50 or younger with at least 3 years of prior job tenure displaced between 1980 and 2005: 77,557 dollars in 2000 dollars over 20 years at a 5 percent discount rate, equal to 1.71 years of predisplacement earnings; 1.4 years when national unemployment was below 6 percent at displacement and 2.8 years when it exceeded 8 percent, with larger losses for men with greater job tenure
   Steven J. Davis and Till von Wachter, Recessions and the Costs of Job Loss, Brookings Papers on Economic Activity, Fall 2011 checked 2026-08-28
9. 09Persistence of earnings losses two decades after displacement Early-1980s displacements produced average annual earnings losses of more than 30 percent of predisplacement earnings relative to a control group, and displaced workers still earned 15 to 20 percent less than the control group implied after 20 years
   von Wachter, Song and Manchester (2011), as reported in Davis and von Wachter, Brookings Papers on Economic Activity, Fall 2011 checked 2026-08-28
10. 10Measured effect of federally funded retraining on displaced workers' earnings The study compared 2,054 TAA participants with 1,796 matched UI claimants laid off in the same period and local areas, average age 49. Training completion was 31 percentage points higher and certificate or degree attainment 30 percentage points higher among participants, yet participants earned 12,674 dollars less in year one and 3,273 dollars less in year four after the unemployment insurance claim. CLEAR rates the causal evidence moderate because the design is nonexperimental
    Schochet, D'Amico, Berk, Dolfin and Wozny (2012), Mathematica Policy Research, as profiled by the US Department of Labor CLEAR evidence review checked 2026-08-28
11. 11CompTIA Security+ SY0-701 exam format and version lifespan Maximum of 90 questions in 90 minutes, passing score 750 on a 100 to 900 scale, launched 7 November 2023, with CompTIA stating versions usually retire about three years after launch, estimated 2026
    CompTIA, Security+ certification page checked 2026-08-28
12. 12Median annual wage for all US workers 50,980 dollars in May 2025
    US Bureau of Labor Statistics, Occupational Outlook Handbook checked 2026-08-28
13. 13Industry, occupation and sex breakdown of long-tenured displaced workers, 2023 to 2025 Nineteen percent lost a job in manufacturing, 16 percent in professional and business services and 10 percent in retail trade. Reemployment was 68.0 percent for men and 64.0 percent for women. By major occupational group the January 2026 reemployment rates ran from 65.2 percent for service occupations to 69.2 percent for sales and office occupations
    US Bureau of Labor Statistics, Worker Displacement: 2023-2025 checked 2026-08-28
14. 14Base counts behind the earnings comparison for long-tenured displaced workers Of the 1.9 million long-tenured displaced workers who lost full-time wage and salary jobs in the 2023-25 period and were reemployed in January 2026, 1.6 million had full-time wage and salary jobs again. Table 7 reports 1,942,000 and 1,593,000 respectively
    US Bureau of Labor Statistics, Worker Displacement: 2023-2025 checked 2026-08-28
15. 15Audit-experiment evidence on how employers read a layoff on a resume, as distinct from a caregiving gap In a randomised audit study of 3,407 real job openings, applicants whose resumes said they had been laid off due to downsizing received interview requests 9.7 percent of the time for mothers and 8.8 percent for fathers, against 15.3 percent and 14.6 percent for otherwise identical continuously employed applicants. The lapse was held at 18 months. No condition in the study added a credential to a resume, so the study says nothing about what a certification does to these rates
    Katherine Weisshaar, From Opt Out to Blocked Out, American Sociological Review 83(1), 2018, free full text hosted by the publisher via the author's open-access eprint link checked 2026-08-28

Related