What is a good employee absence rate?
The U.S. Bureau of Labor Statistics reports an absence rate of roughly 3.0% for private-sector wage and salary workers, and about 3.6% for state and local government workers. A company-wide rate at or below 3% is generally considered healthy. Rates between 3-5% warrant monitoring, and sustained rates above 5% usually signal a workforce or management issue — burnout, low engagement, unsafe conditions, or weak attendance policy enforcement. Benchmark by industry, not just headline number: healthcare and manufacturing typically run higher than professional services.
How do you calculate the absence rate?
Divide lost workdays by total scheduled workdays and multiply by 100. The standard BLS formula is: Absence Rate = (Hours Absent / Hours Usually Worked) x 100, which most HR teams approximate as (Days Lost / Days Scheduled) x 100. For a single employee with 12 absent days out of 240 scheduled days, the rate is 5.0%. For a team, sum every employee's absent days as the numerator and every employee's scheduled days as the denominator. Include full-day absences only; partial-day absences are tracked separately in most HRIS systems.
Should FMLA leave count in the absence rate?
It depends on what you're measuring. For workforce planning and coverage cost analysis, include FMLA leave because the seat is still empty and work still has to be reassigned. For attendance discipline, exclude FMLA-certified absences entirely — the Family and Medical Leave Act prohibits employers from counting protected leave against employees under no-fault attendance policies, and DOL Wage and Hour Division has pursued enforcement actions on exactly this issue. Most HR teams report two absence rates: a total rate and an unprotected rate.
How are paid and unpaid leave treated differently?
Paid leave (PTO, sick days, paid vacation) and unpaid leave are usually tracked in separate buckets even though they're both absences from work. For absenteeism analysis — meaning unplanned, often last-minute time off — most HR teams exclude pre-approved PTO and only count sick days, no-shows, and tardiness that converts to a half-day absence. For workforce cost analysis you'd include both, because paid leave still has a wage cost and unpaid leave still has a coverage cost. SHRM recommends segmenting reports by leave type rather than rolling everything into one number.
What is the Bradford Factor and how does it differ from absence rate?
The Bradford Factor is a UK-origin HR metric that weights short, frequent absences far more heavily than single long absences, on the theory that one 10-day flu is less disruptive than ten 1-day absences. The formula is B = S^2 x D, where S is the number of separate absence spells and D is the total days absent. Ten 1-day spells score 1,000; one 10-day spell scores 10. Absence rate, by contrast, treats all lost days equally. Many UK and increasingly some US employers use both: absence rate for benchmarking and Bradford Factor for triggering attendance review thresholds.
How often should absence rate be calculated?
Monthly is the standard cadence for operational reporting because it's frequent enough to spot trends but smooth enough to avoid noise from a single bad week. Roll up to quarterly and annual for executive dashboards and year-over-year benchmarking. Some HR teams calculate a rolling 12-month rate to dampen seasonality — flu season, summer vacation, and December holidays can swing monthly figures by 1-2 percentage points without indicating any real change in workforce health.
What industries have the highest absence rates?
BLS data consistently shows the highest absence rates in healthcare and social assistance, manufacturing, and public administration, often running 4-5% or higher. Lowest rates are typically in professional and business services and finance, often near 2%. Drivers include physical job demands, shift work, exposure to illness, and union contract protections for sick leave. When benchmarking your company, compare against your NAICS industry code rather than the all-private-sector average.