Skip to main content
Remindax
AI SmartDoc NEW Pricing

Login Signup Free
HR Glossary

Dual Labor Market: What It Means for Employers

A dual labor market is an economic theory that the labor market is split into a primary sector of stable, well-paid jobs with benefits and promotion ladders and a secondary sector of low-paid, unstable jobs with few benefits and little advancement.

  • Last reviewed November 2026
  • 11 min read
  • 21 primary sources
General information - not legal, tax or HR advice

Remindax does not provide legal, tax, or HR advisory services, and this page is for general informational purposes only - consult a qualified attorney, accountant, or HR professional for guidance specific to your situation.

Section 01

Quick facts

  • ✓The dual labor market is a hypothesis in labor economics, not a legal category; eCFR and Federal Register full-text searches found no federal regulation or rule document using the term (checked November 2026).
  • ✓A Federal Reserve staff working paper using CPS data for 1980-2021 estimated that about 55 percent of the population is in a primary sector and 14 percent in a secondary sector whose workers have six times the turnover and ten times the likelihood of unemployment (FEDS 2023-031).
  • ✓March 2026: 39 percent of private industry workers in the lowest-paid 25 percent of occupations had access to employer medical care, versus 94 percent in the highest-paid 25 percent (BLS National Compensation Survey) [VERIFY ANNUALLY].
  • ✓June 2026: private employers spent $5.05 an hour worked on benefits for part-time workers and $17.03 for full-time workers (BLS Employer Costs for Employee Compensation) [VERIFY ANNUALLY].
  • ✓January 2026: median tenure with the current employer was 2.4 years in leisure and hospitality and 5.6 years in the public sector, against 4.1 years overall (BLS Employee Tenure) [VERIFY ANNUALLY].
Section 02

What it means in practice

What the term means. The dual labor market hypothesis holds that jobs fall into two broad tiers rather than one continuous market. A Federal Reserve Board staff working paper (Ahn, Hobijn and Sahin, FEDS 2023-031) summarizes the original idea, which it credits to Doeringer and Piore (1970): primary-tier jobs "generally have low turnover, pay high wages, come with benefits, offer potential for job advancement, and provide job security," while secondary-tier jobs "have high turnover, pay low wages, come with limited benefits, offer few career opportunities, and provide little job security." A 1996 Monthly Labor Review article by BLS economists put it as a "good job" sector and a "bad job" sector, with bad jobs marked by low pay, poor working conditions, few benefits and "poor prospects for moving into the good job sector." The phrase describes how economists model the labor market. It creates no rights or duties, and no employer is legally required to classify its jobs this way.

A contested theory, not settled fact. Government research presents the idea with caveats. A 2006 Monthly Labor Review article on earnings mobility says the theory "was difficult to verify empirically, because the primary and secondary labor segments proved too complicated to separate accurately," faced "sustained criticism as being overly simplistic," and that segmented markets, "if they exist, become irrelevant if workers can move between them." A 1994 BLS working paper (Lettau, Working Paper 260) notes that the explanation, "although often cited," had not received "much direct empirical attention," partly because household surveys describe people rather than jobs. The 2023 Federal Reserve paper says the hypothesis "fell into disfavor among macroeconomists" in the 1980s and 1990s and quotes an early critic calling dual market theories "too varied, incomplete, and amorphous." Its disclaimer says the conclusions are the authors' and do not indicate concurrence by the Board of Governors.

What recent research found. The FEDS paper tested the idea with a statistical model applied to more than 10 million Current Population Survey histories from 1980 to 2021 and added a third, "tertiary" segment of people who participate only occasionally. Primary-sector workers "are almost always employed." The secondary sector was 11.9 percent of employment but accounted for 61 percent of unemployment. In the paper's estimates, median tenure was 5.0 years in the primary sector and 1.8 years in the secondary sector, and median hourly earnings in the primary sector were about 30 percent higher than in the other two tiers. Only about 1 percent of primary-sector workers reported working part time for economic reasons, against about a third in the secondary sector.

Why segments exist: the paper's answer. The authors group proposed causes into five themes: life-cycle career choices, discrimination, insider-outsider effects from unions and other institutions, efficiency wages, and swings in labor demand. They found racial and ethnic minorities and non-naturalized immigrants overrepresented in the secondary tier, but demographic characteristics explained only a small part of who was in which segment (an R-squared of 0.049 for the secondary tier), and that importance declined over time. Union membership was 12.8 percent in the primary tier and 10.3 percent in the secondary tier, so they found unions "not the most important factor." Their evidence fit efficiency-wage explanations best: primary jobs cluster in high-skilled service occupations where effort is hard to monitor, and they pay higher returns to schooling and experience.

How this differs from related terms. The labor market entry covers how supply and demand are measured through JOLTS, the Employment Cost Index and OEWS wage data; the labor force entry covers who BLS counts as employed or unemployed. The dual labor market idea is a claim about the structure inside that market: that it divides into tiers with limited movement between them. Contingent workers are one measurable group often placed in the secondary tier, but the two are not the same thing: a contingent job is defined by whether it is expected to last, while the dual market theory also covers permanent jobs with low pay and no ladder.

Benefits: the clearest divide in BLS data. The National Compensation Survey release of September 25, 2026 (March 2026 data) groups private industry workers by the average wage of their occupation. Among workers in the lowest 25 percent, 39 percent had access to employer medical care and 19 percent participated; in the highest 25 percent, 94 percent had access and 68 percent participated. Retirement access was 48 percent versus 92 percent, and participation 22 percent versus 81 percent. Paid sick leave was 61 percent versus 95 percent and paid vacation 55 percent versus 94 percent. For private industry, the 25th percentile wage was $18.99 an hour and the 75th percentile $40.44. Part-time status shows a similar gap: 23 percent of private industry part-time workers had medical care access against 87 percent of full-time workers. BLS notes that estimates use the average wage for the occupation, so an individual can fall into a different category. For the benefits themselves, see employee benefits.

What benefits cost per hour. The BLS Employer Costs for Employee Compensation release (June 2026 data, last modified September 9, 2026) shows private employers paying part-time workers $25.20 an hour in total compensation, with $5.05 (20.0 percent) in benefits, of which 96 cents was insurance. Full-time workers received $54.00, with $17.03 (31.5 percent) in benefits and $4.58 in insurance. Union workers received $63.06 with 40.5 percent in benefits; nonunion workers $45.49 with 28.7 percent. These are averages across very different jobs. The 1994 BLS working paper, which compared part-time and full-time jobs in the same establishment and occupation, still found compensation per hour "substantially lower" in part-time jobs, with "much lower benefits per hour."

Job stability and underemployment. The BLS Employee Tenure release of September 24, 2026 reported January 2026 median tenure of 4.1 years overall, 5.6 years in the public sector and 3.9 years in the private sector. By industry it was lowest in leisure and hospitality (2.4 years); by occupation, 4.9 years in management, professional and related jobs against 2.9 years in service jobs. Turnover follows the same pattern. The JOLTS release for September 2026 (preliminary, seasonally adjusted) shows a quits rate of 5.0 percent in accommodation and food services and 0.9 percent in finance and insurance, against 2.1 percent for total nonfarm. In October 2026, 4.8 million people worked part time for economic reasons, meaning they wanted full-time work but their hours had been cut or they could not find a full-time job (BLS Employment Situation, released November 6, 2026). Turnover measurement is covered in turnover rate.

Example 1: reading the benefits gap (our arithmetic). From the March 2026 NCS figures, medical care access in the highest-paid quarter (94 percent) is 94 / 39 = about 2.4 times that of the lowest-paid quarter, but participation (68 versus 19 percent) is about 3.6 times higher, because the take-up rate is also lower at the bottom (48 versus 72 percent). Retirement participation, 81 versus 22 percent, is about 3.7 times higher.

Example 2: annual benefit spending (our arithmetic). Applying the June 2026 ECEC averages to a part-time job of 20 hours a week and a full-time job of 40 hours, each for 52 weeks: 1,040 hours x $5.05 = $5,252 in benefits for the part-time job and 2,080 hours x $17.03 = $35,422 for the full-time job. Per hour the gap is $11.98. These are national averages across different jobs.

Federal rules that draw lines by hours. Several federal thresholds turn on hours, so where a job sits relative to them shapes what it carries. Under the ACA employer rules, a full-time employee averages at least 30 hours of service a week (26 CFR 54.4980H-1(a)(21)). Under ERISA, a "year of service" for retirement plan participation is a 12-month period with at least 1,000 hours (29 U.S.C. 1052(a)(3)(A)); IRS Publication 560 (2025) adds that for 401(k) plan years beginning after 2024, a plan generally cannot keep out an employee who has two consecutive 12-month periods of at least 500 hours. FMLA eligibility requires 12 months of employment and 1,250 hours of service in the prior 12 months (29 CFR 825.110). Minimum wage and overtime apply to covered non-exempt employees regardless of hours. Details are on the part-time hours, ACA and minimum wage pages.

Example 3: where three schedules land (our arithmetic). A 32-hour weekly schedule is full-time for the ACA and gives 1,664 hours a year, above the 1,000-hour and 1,250-hour lines. A 28-hour schedule is below the ACA line but gives 1,456 hours, still above both annual lines. An 18-hour schedule gives 936 hours: under 1,000 and 1,250, but over 500, so after two such consecutive years the 401(k) rule above can apply.

Discrimination law and tiered jobs. Because the FEDS paper found minorities overrepresented in the secondary tier, how workers are steered into job tiers matters. Title VII makes it unlawful for an employer "to limit, segregate, or classify" employees or applicants in any way that would deprive them of employment opportunities because of race, color, religion, sex or national origin (42 U.S.C. 2000e-2(a)(2)). A practice with a disparate impact is unlawful unless it is "job related for the position in question and consistent with business necessity," or an available less discriminatory alternative is refused (2000e-2(k)). Different terms under a bona fide seniority or merit system are allowed if not the result of an intention to discriminate (2000e-2(h)). See adverse impact and career path.

Where states differ: two examples. Some states regulate features the theory attributes to secondary jobs. Oregon's Bureau of Labor and Industries says its predictive scheduling law (ORS 653.412 to 653.490) covers retail, hospitality and food service employers with 500 or more employees worldwide. They must post written schedules at least 14 calendar days in advance, and give a written good-faith estimate of hours at hire. Unless the employee agrees, they may not schedule work in the 10 hours after the previous day's shift ends, and hours worked in that window are paid at one and one-half times the regular rate. Without the required notice, an added shift earns one extra hour of pay, and a cut or canceled shift earns half the regular rate for each scheduled hour not worked. Washington's Department of Labor and Industries says paid sick leave applies "regardless of full-time, part-time, temporary, or seasonal status," at least one hour per 40 hours worked, with balances of 40 hours or less carried over. Other states set their own rules.

Example 4: the state rules applied (our arithmetic). A covered Oregon employee paid $18 an hour has a posted 6-hour shift canceled without the required notice: 6 x $18 x 0.5 = $54 in predictability pay. A Washington employee working 20 hours a week for 52 weeks (1,040 hours) accrues at least 1,040 / 40 = 26 hours of paid sick leave.

Using the idea at your own organization. The theory gives no legal test, but the BLS measures above can be compared with your own records: benefit access and participation by pay band against the NCS wage quartiles, tenure and quits by job family against the BLS tenure and JOLTS industry figures, and hours by employee against the 30, 500, 1,000 and 1,250-hour thresholds. FLSA rules require recording hours worked each workday and workweek for non-exempt employees (29 CFR 516.2(a)(7)), which is the data every hours-based test depends on.

FAQ: Is a dual labor market illegal? No law we found prohibits or requires it; it is a description used by economists. Classifying workers into job tiers because of a protected characteristic is prohibited under Title VII, as described above. Is part-time work always secondary-sector work? No. The 1994 BLS paper cites research finding that women in part-time professional and managerial jobs averaged higher hourly wages than similar full-timers. Does the government publish primary and secondary counts? No. The FEDS estimates are research results, not an official BLS series. Do unions explain the split? The FEDS authors found unions only a limited factor; see union.

Section 03

Why it matters and common mistakes

Presenting the dual labor market as an established law of economics or a legal category; BLS and Federal Reserve research describe it as a contested hypothesis.

Treating the FEDS 2023-031 primary and secondary sector shares as official BLS statistics; they are staff research estimates from 1980-2021 data.

Comparing only benefit access and ignoring participation; in March 2026 NCS data, take-up of medical care was 48 percent in the lowest-paid private quarter versus 72 percent in the highest.

Assuming part-time or low-wage workers fall outside federal rules; minimum wage and overtime apply regardless of hours, and the 500-hour 401(k) path reaches some employees under 1,000 hours.

Assigning workers to "core" and "flexible" job tiers in a way that limits, segregates or classifies them by a protected characteristic (42 U.S.C. 2000e-2(a)(2)).

Applying national tier patterns without checking state law, such as Oregon predictive scheduling or Washington paid sick leave for part-time, temporary and seasonal workers.

Section 04

How Remindax helps

Remindax helps you track the dates and deadlines tied to benefit eligibility dates, hours-threshold reviews and schedule notices and sends reminders before they're due.

  • ✓Reminders by Email, SMS and WhatsApp
  • ✓An owner and a due date on every item
  • ✓Forever-free plan, no credit card required
Start forever free
Section 06

Sources & References

Last reviewed: November 2026. Figures are reviewed annually - see Sources & References for current figures.

Keep every HR date on one calendar

Track the deadlines tied to benefit eligibility dates, hours-threshold reviews and schedule notices - with automatic reminders before anything is due.

GDPR-ready | AWS secure cloud | Encrypted storage | Setup under 5 minutes