Quick answers
Bottom line. People didn’t stop wanting to work. Small employers lost the advantage they had in the 2010s, when unemployment was high and workers were plentiful. Now big employers out-pay and out-benefit them, experienced workers are retiring, and hiring is broken for both sides. That’s why owners hear “no one wants to work” while job seekers hear “no one is hiring.” The part you control is your offer and how fast and well you treat candidates. For private practices, the squeeze is tightest for medical assistants and the front desk, where you’re often recruiting against health systems.
- Do people really not want to work anymore? No. (Common misconception) Prime-age (25–54) participation hit 84.0% in July 2024 and again in January 2026, the highest since 2001, against 82.9% before the pandemic (BLS LNS11300060). Headline participation is down because the population is older.
- What’s different for private practices? Demand is growing and the competitor is a system. Physician-office employment hit 3.06 million in September 2026, up 12.4% from February 2020 (BLS CES6562110001). Practices recruit against hospitals, where 95% of workers have a retirement plan (BLS USDL-25-1464).
- Is this new? No. The phrase appears in an April 1894 Kansas newspaper during a coal strike and recurs in booms and busts alike (Snopes).
- Is there a labor shortage, or is no one hiring? Neither, nationally. In August 2026 there were 7.08 million openings and 7.03 million unemployed people, about one job per jobseeker, down from 2.04 in March 2022 (BLS JOLTS openings; BLS unemployed). What’s unusual is how little moves: quits are below their 2019 rate (BLS quits rate).
- Why are talented people ghosted and out of work? Low churn traps outsiders. 27.1% of the unemployed have looked for six months or more, up from 23.6% a year earlier (BLS LNS13025703). Recent graduates are 5.6% unemployed and 42% underemployed (NY Fed). 61% of jobseekers report being ghosted after an interview (Greenhouse 2024 (vendor survey)).
- What changed since the 2010s? Owners’ bargaining edge. Unemployment averaged 7.4% to 9.6% a year from 2010 to 2013 (BLS LNS14000000). Since 2019, real pay at the 10th percentile rose 15.3% against 5.8% at the median (EPI), and big-employer pay floors moved far above the legal minimum.
- Why can’t small businesses compete with big employers? Total compensation. Amazon’s floor is $20 an hour (FashionNetwork/Reuters). In 2026, 66% of firms with 10–199 workers offer health coverage against 98% of larger firms (KFF 2026).
- Is the population aging making it worse? Yes, permanently. The 1.7-point age-mix drag works out to about 4.7 million fewer workers than a 2019 age mix would give (our calculation from St. Louis Fed and BLS CPS population). Net migration in 2025 was probably negative for the first time in at least 50 years (Brookings).
- Did stimulus checks make people lazy? No. (Common misconception) Where the $300 unemployment supplement ended early, workers lost $2,529 in benefits on average and gained $292 in earnings (Coombs et al., AEA P&P 2022). Either way, the programs ended in 2021.
- Is Gen Z lazy? No; they’re less experienced. (Common misconception) A meta-analysis of 20 studies found generational differences in work attitudes “moderate to small, essentially zero in many cases” (Costanza et al. 2012). What fell is early work experience: teen summer employment dropped from 58% in 1978 to 35.5% in 2026 (Pew).
- Is consolidation squeezing small businesses? Yes, slowly, and fastest in healthcare. Small businesses employ 45.9% of private workers but pay 38.7% of private payroll (SBA Advocacy 2026). 82.0% of physicians now work for hospitals or corporate entities (PAI–Avalere 2026).
- Is AI taking jobs? At the young, exposed edge, not across the economy. Employers cited AI for about 21% of announced cuts through September 2026 (Challenger, Sep 2026). One payroll study puts employment of 22–25-year-olds in the most AI-exposed jobs 19% below trend (Brynjolfsson, Chandar & Chen, Aug 2026 version), but Yale’s Budget Lab sees no economy-wide disruption yet (Yale Budget Lab).
- What should I do about it? Compete on the offer and the process. Post the pay, reply fast, decide within a week, sell total compensation and stable schedules, and onboard deliberately. Small employers can do most of this faster than large ones.
Every figure below links to the source a fact-check validated on October 22, 2026. Vendor surveys are labelled. Caveats and dropped figures are in the appendix.
The complaint is old, and it reliably flatters the person making it
Political scientist Paul Fairie’s viral July 2022 thread collected “nobody wants to work” clippings from 1894 to 2022 (InsideHook; Snopes). In 1937, amid mass unemployment, a Pennsylvania peach grower said nobody wanted to pick fruit, which can only be a complaint about the wage (Ag Proud). Georgetown historian Joseph McCartin says shortages often reflect jobs that “just don’t pay enough” (FA-Mag/Bloomberg).
The 2021 wave had a policy target. A McDonald’s sign reading “Nobody wants to work anymore” went viral in April 2021, and the person who posted it told Fox News that people made more on unemployment (Fox News). In May 2021, 76% of Republicans and 24% of Democrats blamed government payments for unemployment, but only 25% of the unemployed did (YouGov). The people who actually quit in 2021 cited low pay (63%), no advancement (63%) and feeling disrespected (57%) (Pew).
Psychology explains why the frame survives. In five preregistered studies, Protzko and Schooler found that adults see “kids these days” as declining on the very traits where they themselves excel: well-read people think youth read less, and authoritarian people think youth are disrespectful. When the researchers gave people false feedback about their reading ability, it changed their memories of how much they read as children. It did not directly change their judgments of today’s youth, only slightly through that memory shift (Protzko & Schooler, Science Advances 2019). Owners who built businesses on long hours are the people this bias predicts will see a work-ethic collapse.
2019 to 2026: a real shortage that fully reversed
In 2021–22 labor really was scarce. Openings peaked at 12.3 million in March 2022, and quits at 3.0% in November 2021 (BLS JOLTS openings; BLS quits rate). Small firms felt it most: 51% of NFIB owners had openings they couldn’t fill in September 2021, a 48-year record (NFIB, Oct 2021). By 2026 every national measure had normalized or overshot.
| Measure | Feb 2020 | 2021–22 high | Latest (Aug–Sep 2026) |
|---|---|---|---|
| Job openings | 6.97M | 12.30M (Mar 2022) | 7.08M (Aug) |
| Openings per unemployed person | 1.21 | 2.04 (Mar 2022) | 1.01 (Aug) |
| Quits rate | 2.3% | 3.0% (Nov 2021) | 1.9% (Aug) |
| Hires rate | 3.9% | 4.6% (Nov 2021) | 3.3% (Aug) |
| Unemployment rate | 3.5% | — | 4.2% (Sep; peak 4.5%, Nov 2025) |
| Prime-age employment rate | 80.4% | — | 80.7% (Sep; peak 80.9%, 2023–24) |
| NFIB owners with unfilled openings | — | 51% (Sep 2021) | 32% (Sep; long-run average 24%) |
Sources: BLS JOLTS openings, BLS unemployed, BLS quits, BLS hires, BLS unemployment rate, BLS LNS12300060, NFIB Sep 2021, NFIB Sep 2026. The ratio is calculated from the two BLS series for the same month.
Pay moved the way a “shortage at the offered wage” predicts. Average hourly earnings rose 32%, from $28.54 in February 2020 to $37.81 in September 2026 (BLS CES0500000003). That has stalled. The 10th-percentile real wage fell 0.3% in 2025 (EPI), real average hourly earnings fell 0.7% in the year to September 2026 as prices rose 3.7% (BLS Real Earnings), and wage growth slowed to 3.9% in September from 4.1% in August (Atlanta Fed Wage Growth Tracker).
Small owners still report more hiring pain than usual. In September 2026, 32% had unfilled openings, and labor quality or availability was the top problem for 26% of owners, against a 12% average. 87% of owners who were hiring, or 45% of all owners, saw few or no qualified applicants. Only 6% named labor cost (NFIB Sep 2026). The 2021 complaint was “no applicants.” The 2026 one is “no qualified applicants,” and owners aren’t raising wages to fix it.
Two markets: insiders are safe, outsiders wait
The hires rate of 3.3% matches the slow early-2010s recovery, when it ran 3.0–3.5% (BLS hires). People with jobs keep them, and people without them wait. The pay premium for switching jobs has shrunk to 1.3 points (4.7% for switchers vs. 3.4% for stayers), from 2.8 in August 2022 (FRED, Atlanta Fed switchers; FRED, Atlanta Fed stayers), so experienced workers rarely enter the applicant pool.
The pain falls on outsiders. In September 2026, 1.9 million people had been unemployed for 27 weeks or more, and 4.5 million worked part time because they couldn’t get full-time hours (BLS Employment Situation, Sep 2026). Bachelor’s holders aged 25 and up had 2.5% unemployment (same release), against 5.6% for recent graduates (NY Fed).
The split by industry is sharp. Information and finance post openings at rates of 4.3% and 5.0% but hire at only 1.6% and 1.7% a month (BLS JOLTS, Aug 2026). Indeed’s software-development postings index sat at 76.6 in September 2026, about 23% below February 2020 (FRED). Health care and social assistance has the most openings, 1.36 million (same JOLTS release). Many need licenses a laid-off analyst lacks. Small-business openings skew skilled too: 27% of NFIB owners had skilled openings against 16% unskilled in September (NFIB Sep 2026).
Announced layoffs are loud but falling. Challenger counted 573,195 through September 2026, down 39% from 2025. Excluding government, they are down only 15% (Challenger, Sep 2026). About 317,000 people left the federal workforce in 2025 (FedManager), under 0.2% of payroll jobs (our arithmetic from BLS CES).
Private practices: a tighter version of the same squeeze
Healthcare is the one sector still adding jobs. Ambulatory care employment is up 17.8% since February 2020 (BLS CES6562100001). Health care and social assistance has an openings rate of 5.4% against 4.3% economy-wide (BLS JOLTS health care), and its quits rate of 2.0% is back to the 2019 level (BLS). Health workers reporting frequent burnout rose from 32% in 2018 to 46% in 2022 (CDC Vital Signs).
The pinch point is medical assistants and the front desk. MA pay rose 4.3% in 2025 and 20.6% over five years (Physicians Practice, citing MGMA). In May 2025, 47% of practice leaders named MAs as the hardest role to recruit (MGMA Stat). In May 2026, 28% reported higher turnover, and MGMA warned that “turnover measures who left. It does not measure whether a practice is fully staffed” (MGMA Stat). 68% of practices added or expanded AI in 2025, mostly for clinical documentation and next for scheduling and patient communications (MGMA Stat).
What changed since the 2010s
Big employers raised the bar on pay and benefits
The floor that matters is no longer the legal minimum. Amazon pays at least $20 an hour, with average total compensation above $32 (FashionNetwork/Reuters), and Costco starts at $20 (RetailWire). In California, health systems with 10,000 or more staff must pay covered workers, including billing and housekeeping, $25 an hour from July 2026 (California DIR). Legal minimums bind every employer; the squeeze comes from large firms paying well above them.
Benefits are where the gap is widest. Health coverage is offered by 66% of firms with 10–199 workers and 98% of larger ones (KFF 2026). Retirement-plan access is 55% at establishments with 1–49 workers against 90% at those with 500 or more (BLS, Mar 2025); the March 2026 data put large establishments at 91% (BLS, Mar 2026). The gaps predate 2020; what changed is that workers could act on them.
The experienced workforce aged out and immigration reversed
The pandemic pulled forward roughly 2.1 to 2.95 million retirements (Federal Reserve Board; St. Louis Fed). That bulge unwound by late 2024, but ordinary aging didn’t: 20.0% of the population was retired in May 2025, against 15.3% in 1995 (St. Louis Fed, Jul 2025). The people leaving are the most experienced, which plausibly feeds “no qualified applicants” (an inference, not a measured link). CBO projects deaths will exceed births from 2030 (Becker’s on CBO).
Immigration helped and then reversed. CBO estimated net immigration of 3.3 million in each of 2023 and 2024 (Boston Fed), which the San Francisco Fed credits with about one-fifth of the easing of labor-market tightness in 2023 (SF Fed). For 2025, Brookings and AEI economists estimate net migration of −295,000 to −10,000; CBO puts it near +400,000 (Brookings).
Hiring broke on both sides
Applying got nearly free. LinkedIn reported about 11,000 applications a minute in mid-2025, up 45% in a year (eWeek on LinkedIn data (vendor)), and US applicants per opening have doubled since spring 2022 (HR Dive on LinkedIn (vendor)). Only about 4% of some 6,500 companies using Greenhouse contacted every rejected applicant in Q2 2024 (Axios on Greenhouse (vendor)), and 34% of class-of-2026 graduates surveyed had an accepted offer pulled (Resume Builder (vendor survey)).
Candidates ghost back. In Indeed’s 2019 survey, 83% of employers had been ghosted, and 40% of the candidates who did it had another offer (Indeed). Filters add waste: 88% of employers said qualified candidates were screened out for not matching criteria exactly (HR Dive on HBS/Accenture).
Consolidation pulls jobs and pay toward scale, most of all in healthcare
Small businesses number 36.2 million but get a smaller share of pay than of jobs: 45.9% of private employment, 38.7% of private payroll (SBA Advocacy 2026). The Chicago Fed finds a “consistent downward trend” in the employment share of firms with fewer than 50 workers, and the larger firms’ share “resumed its upward trajectory” after the pandemic (Chicago Fed Letter 509). Over 75% of US industries have grown more concentrated since the late 1990s (Grullon, Larkin & Michaely), though locally the pressure is usually a national chain arriving in town (NBER w25066). In local labor markets, moving from the 25th to the 75th percentile of employer concentration goes with 5% to 17% lower posted wages (Azar, Marinescu & Steinbaum, JHR).
Healthcare is the extreme case. As of January 1, 2026, 82.0% of physicians were employed by hospitals (59.7%) or corporate entities (22.3%), up from 77.6% two years earlier, and 63.9% of practices were hospital or corporate owned (PAI–Avalere 2026). Fewer than one in five physicians works in a physician-owned practice (Medical Economics). Private equity raises prices after buying in: charges in dermatology, gastroenterology and ophthalmology rose 20.2% (NIHCM on Singh et al.), and anesthesia prices rose 26% at PE-backed groups (JAMA Internal Medicine).
Two policy changes matter for independents. The FTC dropped its national non-compete ban in September 2025, so enforceability is mostly back to state law, with some case-by-case FTC enforcement (Morgan Lewis). Medicare’s 2026 conversion factor rose 3.26%, but most of that is a one-year 2.5% boost. A separate −2.5% “efficiency adjustment” cuts work RVUs for non-time-based services; E/M visits are exempt, and CMS will recalculate it on a three-year cycle (Holland & Knight).
Young hires: a gap in experience, not character
The workplace evidence doesn’t show a generational character change (Costanza et al. 2012). It does show a real experience gap. Teen labor-force participation fell from 57.9% in 1979 to 34.1% in 2011, mostly because of school enrollment (BLS Monthly Labor Review). Young adults do work, just later: in July 2026, 67.5% of 20–24-year-olds were employed, against 62.4% in July 1948 (American Action Forum). Habits once learned in a teenage job, like punctuality and handling customers, now get taught at 22.
Schooling was disrupted. The 2024 12th-grade NAEP produced the lowest math and reading averages ever reported, with 45% of seniors below Basic in math (NAGB). Chronic absenteeism peaked at 28.5% in 2022, against about 15% before the pandemic, and was still about 22.6% in 2024–25 (Chalkbeat on AEI data). Persistent sadness among high schoolers went from 28.5% in 2011 to 42% in 2021 (CDC MMWR), then fell to 33.3% in 2025 (CDC MMWR 2026). The role of phones is contested; one large study found technology use explains at most 0.4% of the variation in teen well-being (Oxford Internet Institute).
What fell most since 2020 is what managers control. Engagement among Gen Z and younger millennials fell from 40% to 35% between March 2020 and November 2023, led by feeling cared about, development and progress conversations (Gallup). Across all ages, only 46% clearly knew what was expected of them in 2024, down from 56% (Gallup). Older millennials fell as much (Gallup): a workplace trend, not a generational trait.
What works for practices and other small employers
None of this comes from randomized trials with small employers. It’s large-firm, platform and state-policy evidence applied by inference. Start by measuring the funnel, from applications to 90-day retention, to see whether the problem is flow or drop-off.
- Post the pay range. After Colorado required pay in job postings, workers’ actual pay grew about 1.3% faster than in other states (NBER w34480). By May 2025, 59% of US Indeed postings showed pay (Indeed).
- Move fast and close the loop. Reply to every applicant within a day or two, phone-screen instead of long forms, state your timeline, offer within a week and never pull an accepted offer. Speed is a small employer’s cheapest advantage.
- Assess skills, not credentials. Large firms that publicly dropped degree requirements changed fewer than 1 in 700 hires. Gains came only from firms that changed how they actually assess people (Burning Glass Institute). Use a work sample, a role-play or a paid shadow shift.
- Sell total compensation and stability. If you can’t match a $20 floor, quote the full value of the package, and weigh whether a health stipend or retirement plan beats another dollar an hour. Schedule instability is a strong predictor of turnover among hourly workers (Shift Project), so a clinic with fixed hours should say so in the ad.
- Manage and onboard deliberately. Managers account for at least 70% of the variance in team engagement (Gallup). A meta-analysis of newcomer research ties retention to role clarity, social acceptance, task mastery and fit (Bauer et al., Journal of Management). Write down the quality standard with examples, hold a short weekly check-in for the first 90 days, name a go-to person on day one and set a clear phone rule.
- Use the hiring window. 1.9 million long-term unemployed people, 4.5 million involuntary part-timers (BLS) and underemployed graduates are a deep pool for front-desk, billing, scheduling and operations roles, though not licensed ones.
- For practices, compete where scale can’t. Offer speed, a predictable schedule, a path from front desk to MA and training for uncredentialed hires. Where state law limits non-competes, freedom from one is a selling point.
Ranked verdict
Rank weighs evidence, effect size and 2026 relevance.
| Rank | Theory | Verdict | Explains | Still relevant in 2026? |
|---|---|---|---|---|
| 1 | Pay and benefits gap against large employers | Strong | Owner | Yes, structural |
| 2 | Frozen, low-churn market | Strong | Both | Yes, the defining feature |
| 3 | Aging and retirements | Strong | Owner | Yes, permanent |
| 4 | Immigration swings | Strong; estimates conflict | Owner | Yes, now a drag |
| 5 | Broken hiring: application flood, slow replies, ghosting both ways | Moderate; mostly vendor data | Both | Yes, and fixable |
| 6 | Skills and licensing mismatch | Moderate | Both | Yes |
| 7 | Consolidation and scale competition | Strong trend; hiring link inferred | Owner | Yes, severe for practices |
| 8 | Less early work experience, disrupted schooling | Strong facts; weak link to performance | Owner | Yes, for entry-level roles |
| 9 | Management, engagement, unstable schedules | Moderate | Owner | Yes, a retention lever |
| 10 | Cost of living and childcare | Moderate | Owner | Yes, rising again |
| 11 | AI cutting entry-level white-collar hiring | Contested | Jobseeker | Rising; cuts demand, not supply |
| 12 | Healthcare burnout | Moderate | Owner | Yes, for practices |
| 13 | Tight 2021–22 market and job-switching | Strong for its time | Owner | No; quits below 2019 |
| 14 | Stimulus, expanded unemployment, excess savings | Real but small | Owner | No; all ended |
| 15 | Generational laziness | Not supported; perception bias | Owner | No |
Alternative takes
| Frame | Usual lean | Strongest point | Rebuttal |
|---|---|---|---|
| Men without work (Eberstadt) | Center-right | 6.8 million prime-age men were out of the labor force in 2024 (BLS CPS), and Eberstadt calls disability programs “perverse” (NYRB) | Health is central: 43% report fair or poor health and 40% say pain prevents full-time work (Krueger, BPEA 2017). A long-run trend about people outside the applicant pool |
| Work doesn’t pay; the “true” unemployment rate | Left and labor | LISEP’s rate, which counts poverty-wage workers, was 24.0% in September 2026 against 4.2% official (LISEP) | It measures job quality, not joblessness, and low-wage workers had the fastest real gains in decades in 2019–24 (EPI) |
| Vibecession and doubts about the data | Mixed | The GAO found large revisions and falling survey response threaten jobs-report quality (GAO-26-107538), and the BLS commissioner was fired after a 258,000 downward revision (Bloomberg Law) | Underfunding, not tampering (PIMCO). Non-BLS sources (NFIB, Challenger, ADP payroll studies) point the same way |
| The skills gap is an employer failure (Cappelli) | Centrist, management scholarship | Employers want fully formed hires they never have to train (SHRM on Cappelli). Only 6% of owners call labor cost their top problem (NFIB Sep 2026), so wages don’t signal a broad shortage | Where shortages are real, wages do rise: MA pay is up 20.6% in five years (Physicians Practice/MGMA) |
| Baumol and the cost of living | Cross-ideological | Wages are 60–70% of childcare providers’ costs, so care can’t get cheaper (Cato); childcare averaged $13,184 in 2025 (Child Care Aware) | Childcare prices rose about as fast as overall inflation since 2021 (same source). A price problem, not a labor-supply one |
| AI is killing entry-level jobs | Tech commentary, jobseekers | 22–25-year-olds in the most exposed jobs are 19% below trend (Brynjolfsson et al., Aug 2026) | No economy-wide disruption yet (Yale Budget Lab); Sam Altman says some firms are “AI washing” layoffs (ITPro) |
Quotable stats
| Stat | Source |
|---|---|
| “Nobody wants to work” appears in US newspapers as early as 1894 | FA-Mag/Bloomberg |
| Prime-age participation peaked at 84.0% (July 2024, Jan 2026), above 82.9% before the pandemic | BLS LNS11300060 |
| 7.08M openings vs. 7.03M unemployed in August 2026: about one job per jobseeker | BLS openings, BLS unemployed |
| 27.1% of the unemployed have looked six months or more, up from 23.6% a year earlier | BLS LNS13025703 |
| 32% of small businesses have unfilled openings (24% average); 87% of those hiring see few or no qualified applicants | NFIB Sep 2026 |
| Small businesses employ 45.9% of private workers but pay 38.7% of private payroll | SBA Advocacy 2026 |
| 66% of small firms (10–199 workers) offer health coverage vs. 98% of larger firms | KFF 2026 |
| The age mix alone cut participation 1.7 points from June 2019 to June 2026 | St. Louis Fed |
| Ending the $300 benefit early: workers lost $2,529 in benefits and gained $292 in earnings | Coombs et al., AEA P&P 2022 |
| 63% of 2021 quitters cited low pay; 57% felt disrespected | Pew |
| Teen summer employment: 58% in 1978, 35.5% in 2026 | Pew |
| Recent grads: 5.6% unemployed, 42% underemployed | NY Fed |
| 82.0% of physicians work for hospitals or corporate entities, up 4.4 points in two years | PAI–Avalere 2026 |
Appendix: caveats and data notes
- Timing and breaks. JOLTS is August 2026; the household survey is September. The openings ratio uses August for both. NFIB renamed its labor-quality category in June 2026, so don’t compare “top problem” readings across it. Pew’s 2026 teen figure covers June–July only.
- Derived figures. “About 4.7 million fewer workers” is 1.7 points times the 2026 population aged 16 and up. The federal share of payroll jobs is also our arithmetic.
- Labels. Ghosting, application-volume and rescinded-offer figures are vendor surveys. PAI is physician-advocacy aligned, and LISEP’s rate isn’t comparable with BLS rates.
- Careful phrasing. Immigration explains about a fifth of the 2023 easing, not most of it. Protzko and Schooler showed a memory bias, not that views of youth can be steered directly. The Canaries estimate has grown with each version (13% on July 2025 data, 16%, then 19% in August 2026). Challenger’s −39% is inflated by 2025 federal cuts.
- Dropped. Quotes that couldn’t be found (Andy Challenger on “the money for those roles,” Futurity’s “probably more complicated”); unverifiable or vendor-only claims (Gen Z “fired” surveys, 78% Gen Z ghosting, NACE 50% vs. 84.6%, RN pay −8.9%, job-hugging and ghost-job rates, Indeed “3.8x applications”); uncited ones (a 2025 drop of 400,000 working women, sector immigration losses); superseded ones (long COVID “2–4 million,” now about 420,000; KFF 59% vs. 97%); the “re-accelerating wages” framing; and the contested FT conscientiousness chart.