• HŪMNZ Newsletter
  • Posts
  • The HŪMNZ Element: Issue 27 - Why Your Retention Numbers Look Fine and Aren't

The HŪMNZ Element: Issue 27 - Why Your Retention Numbers Look Fine and Aren't

The national quits rate has not moved in a year. Underneath it, your knowledge workers are staying because they cannot leave, and your frontline is still walking out.

Author

Tim Dennis
August 20, 2026

💡 Editor's Note

If your turnover is flat this year, the temptation is to read it as a win.

Before you do, it is worth asking what a flat number is actually made of.

Are our people staying, or are they stuck?

Those two conditions produce an identical retention report. They behave completely differently the moment the labour market moves.

Executive Brief

Bottom line: A stable company-wide turnover rate is not evidence of retention health, because it averages together two opposite problems, knowledge workers who are staying involuntarily and frontline roles that are still churning at near-record rates.

The national picture has barely moved. According to the US Bureau of Labor Statistics, in June 2026 job openings stood at 7.4 million, hires at 5.3 million, and quits at 3.2 million — a quits rate of 2.0%. The BLS described openings, hires, separations and quits as little changed or unchanged.

Stability at the top hides a split underneath it. The Burning Glass Institute, analysing quits by segment, found a two-track labour market: workers with advanced degrees quitting at near-historic lows, while several frontline occupations sat in the top quartile of their own 25-year range.

One number. Two problems pointing in opposite directions. Neither one visible in a company-wide average.

In 30 seconds

  • The US quits rate was 2.0% in June 2026 and the BLS called it unchanged, meaning a flat turnover number this year reflects the market, not necessarily your management.

  • The Burning Glass Institute found PhD holders quitting at the 3rd percentile of their 25-year range and computer and mathematical occupations at the 20th, while protective services and installation, maintenance and repair sat at the 87th.

  • Stop reporting one turnover rate. Segment voluntary exits by job family, wage band and tenure, and separate regretted from non-regretted departures before the next review.

⚠️ Four signals leaders should be watching this week

1. Why does a stable turnover rate hide two opposite problems?

Signal: Company-wide retention looks steady while individual managers report very different experiences of the same year.

Evidence: The BLS reported a US quits rate of 2.0% for June 2026, unchanged from the prior month. The Burning Glass Institute's segmented analysis found that same aggregate concealed a two-track market — advanced-degree and white-collar quits at historic lows, several blue-collar occupations in the top quartile of their own 25-year range.

Implication: An average is only useful when the populations inside it behave the same way. Yours do not.

Action: Split your voluntary exit reporting by job family and wage band before your next operating review. If the two halves move in opposite directions, you never had one retention problem.

2. What does loyalty look like when it is really immobility?

Signal: Regretted attrition among senior and technical staff has fallen, and nobody can explain what changed internally to cause it.

Evidence: The Burning Glass Institute found quits among PhD holders at roughly the 3rd percentile of the past 25 years, master's holders similarly depressed, and life, physical and social science occupations at the 11th percentile. Workers under 40 showed the largest declines of any age group.

Implication: If nothing you did explains the improvement, nothing you do will protect it. Suppressed mobility is a market condition on loan, not a retention strategy.

Action: Ask your top quartile of performers what would make them take a call from a recruiter. Do it while they are still answering honestly.

3. Where is churn still costing you real money right now?

Signal: Frontline and operational roles keep turning over while leadership attention sits on knowledge-worker retention.

Evidence: The Burning Glass Institute placed protective services and installation, maintenance and repair at the 87th percentile of their 25-year quits range, healthcare support at the 80th and healthcare practitioners at the 77th. SHRM has reported that replacing an employee can cost between 50% and 200% of their annual earnings.

Implication: The roles closest to delivery and customer experience are the ones still walking, and their replacement cost is rarely modelled with the same care as a senior hire.

Action: Calculate fully loaded replacement cost for your three highest-turnover frontline roles. Compare it to what you spend retaining them.

4. What should you measure instead of one retention number?

Signal: The retention metric on the executive dashboard cannot tell you whether anyone chose to stay.

Evidence: Gallup's State of the Global Workplace 2026 reports that only 20% of employees worldwide were engaged at work in 2025. Engagement that low alongside historically low quits describes a workforce that is present without being committed.

Implication: Turnover is a lagging indicator of a decision employees made months earlier. Engagement and intent are the leading ones, and most operating reviews carry neither.

Action: Add three measures to the review: voluntary exits split regretted versus non-regretted, exits by manager, and internal mobility as a share of open roles filled.

Stat of the Week

2.0%

The US quits rate in June 2026, which the Bureau of Labor Statistics described as unchanged — 3.2 million people leaving jobs voluntarily out of a workforce where 7.4 million roles sat open.

A number that has not moved in a year is not telling you your retention improved. It is telling you the market went quiet.

Source: US Bureau of Labor Statistics, Job Openings and Labor Turnover Survey, June 2026. Segment percentiles from the Burning Glass Institute's analysis of quits by education and occupation.

The Core Question

Every retention report answers the same question: how many people left?

Almost none of them answer the question that actually predicts next year.

Are our people staying, or are they stuck?

The VALŪE lens: Retention measures whether someone left. It does not measure whether they chose not to. Those are different facts with different futures, and only one of them survives a hiring recovery.

The organisations that will hold their best people through the next upturn are the ones treating this period as borrowed time rather than earned loyalty.

Until next time,

The HŪMNZ Element — Weekly Pulse

If this was useful, forward it to one person: a CEO who is about to read a flat turnover report as good news, or the one member of your leadership team who owns retention.