Every late October, the same scene repeats in the parking lots of America's distribution centers: staffing-agency buses rolling in, "Now Hiring for Peak Season" signs by the gate. Nearly every warehouse manager faces the same question — order volume is about to double or triple, where do the people come from? The instinctive first move is a spreadsheet comparing hourly rates: temp labor bills around $28 an hour, a full-time employee costs roughly $25 all-in. Temps are a bit pricier but you don't carry them year-round. Case closed, or so the spreadsheet suggests.
The spreadsheet is wrong in a very instructive way. It equates labor cost with the hourly rate, and the hourly rate is only the most visible — and least complete — piece of the puzzle. According to publicly published rate guides from staffing firms, the markup on warehouse temp labor typically runs 25% to 60% above the worker's pay. More fundamentally, a bill rate buys a warm body on the floor, not a unit of output: a new hire learning your slotting logic from zero can produce at half the units-per-hour of a two-year veteran, and that gap flows straight into the labor cost of every order shipped. The right question was never "whose hourly rate is lower." It is "whose cost per unit shipped is lower" — and answering it requires opening four separate ledgers.
Ledger one: the temp-labor agency markup, a hidden tax in the invoice
Public rate guides place the warehouse-temp markup between 25% and 60% — large, steady commitments price toward the bottom, last-minute peak requests toward the top.
That markup is not pure profit — it covers payroll taxes, workers' compensation, recruiting, and overhead. But from the warehouse's side of the table, it functions as a channel tax: the identical picking motion costs a quarter to three-fifths more through an agency. Benchmark against the bill rate, not the worker's pay.
Ledger two: the productivity gap — who pays for the learning curve
A new hire typically needs several weeks to reach 80% of a veteran's UPH. For full-timers, that learning period happens off-season and amortizes across a year of orders. For temp labor, it lands in the weeks when orders are most expensive and most urgent.
A post-peak review from a ~150,000 sq ft e-commerce fulfillment center in California's Inland Empire (Q4 2024) illustrates the point. Daily orders surged from 12,000 to 35,000; temps peaked at 180 against 120 full-timers. Temps averaged ~60% of veteran UPH in their first two weeks, ~80% by week four, with error rates two to three times the full-time level — and the mis-ships consumed three full weeks of returns triage after the peak. Those three weeks appeared in no peak-season budget.
There is also a second-order effect that rarely gets measured. During Black Friday week in 2023, a home-goods retailer's Midwest warehouse pulled veterans off replenishment to protect pick throughput. Replenishment fell behind, pickers hit empty slots in week two, and productivity slid anyway. The temp gap was filled by cannibalizing the full-time team — a hidden cost that never shows up on the temp ledger.
Ledger three: training and safety — the risk premium on temp labor
Under OSHA's framework, agencies and host employers share responsibility for temp-worker safety — the agency covering general training, the host covering site-specific hazards, PPE, and supervision. The responsibility is joint, but the cost of an incident — downtime, investigation, claims, premium increases — lands overwhelmingly on the warehouse.
Occupational-safety research has documented the pattern for years: temporary workers file injury claims at significantly higher rates than permanent employees in the same industries with comparable tenure. Researcher Michael Foley's matched comparison found this cannot be explained by temps doing more dangerous work — the gap comes from preparation: less screening, less safety training, less control over their own tasks. The expected accident cost per labor hour is simply higher for a temp population.
The training side of this ledger is equally unforgiving: train a full-timer once and the investment pays off for years; train a temp cohort and the residual value hits zero when the assignment ends.
Ledger four: conversion and attrition — where temps go after peak
Many warehouses treat temp staffing as outsourced probation: convert the good ones. But conversion has a price — temp-to-perm fees run as a percentage of annual salary, highest after short assignments, and a temp who proved herself through peak usually wants a raise. The alternative is no cheaper: release the cohort, and next peak you recruit and train from zero. The loop is familiar — pay to keep people, or pay the "newcomer tax" every year. Full-time logic runs in reverse: recruiting and training are one-time investments, and every year of tenure raises the return per labor hour.
Adding it up: temp labor vs full-time cost per unit
Put the four ledgers together and the comparison shifts from the hour to the order. The table below is an illustrative model built on public markup ranges and industry rules of thumb:
| Cost component | Full-time (year-round) | Temp (12-week peak) |
|---|---|---|
| Nominal / bill rate | ~$20/hr wage, ~$25/hr employer cost | ~$28–32/hr bill rate (25–60% markup) |
| Steady-state UPH | 100 (baseline) | ~60–80 in month one, ~90 in month two |
| Effective labor cost per unit | ~$0.25 | ~$0.38–0.45 |
| Training investment | One-time, amortized yearly | Repeated per cohort, zero residual |
| Errors and rework | Low | 2–3x higher, plus post-peak triage |
| Expected safety cost | Low | Significantly higher claim rates |
The model points to a counterintuitive conclusion: temp labor's hourly rate runs only 20–30% above full-time, but its cost per unit shipped can run 50% or more above. What costs extra is not the person — it is the newness: the productivity discount, the errors, the risk, the annual ritual of becoming new again.
A decision framework: three peak shapes, three staffing structures
Once the costs are visible, the decision becomes a matching problem: match the staffing structure to the shape of your peak.
The first shape is the sharp spike: high and short — Black Friday and Cyber Monday at four to five times normal volume for two weeks, then back to baseline. Temp labor is the rational choice here, with two conditions: start recruiting and training at least six weeks early; and keep temps on the simplest, lowest-error-cost tasks (packing, labeling), with picking and exceptions reserved for veterans.
The second shape is the long ramp: two to three months at roughly twice normal volume — the whole of Q4. Here the temp cost-per-unit disadvantage compounds with time. The better structure is a full-time core plus conversion-track temps: hire full-timers ahead of the season, season them off-peak, layer temps on top for elasticity — and convert the strong performers, so training compounds instead of evaporating.
The third shape is the elevated baseline: nominally a "peak season," but volume climbs year after year and the peak is becoming the new normal. Temp-heavy staffing here means using a short-term tool for a long-term condition and repaying the newcomer tax annually. The correct move is to re-baseline headcount and convert part of the "seasonal" roles to permanent ones.
The trend: the billing unit is shifting from heads to units
Three background trends are worth watching.
First, labor-market tightness looks structural. US warehouse turnover has long run above the manufacturing average, peak recruiting gets harder yearly, and the top of the markup range keeps rising. As "being able to hire at all" becomes the scarce resource, the hidden costs — longer recruiting cycles, unstable show-up rates, compressed training — keep rising.
Second, automation is redefining the worker. Over the past five years, AMRs, sort walls, and goods-to-person systems have spread rapidly through mid-size and large warehouses. Machines have no learning curve and no attrition. As basic picking moves to machines, remaining human roles concentrate in exception handling, equipment collaboration, and quality control — the tasks where temporary workers are weakest.
Third, the billing model itself is loosening. Some 3PLs have begun experimenting with per-unit pricing for outsourced labor: pay per order produced, regardless of headcount. That transfers efficiency risk to the staffing provider — and forces everyone to use cost-per-unit as the yardstick. Once that yardstick is universal, the temp-versus-full-time debate resolves itself.
Which leaves an open question worth sitting with: if the unit of account in the warehouse truly shifts from heads to units — labor billed per order, machines depreciated per order — does the temp-versus-full-time debate become a pseudo-problem? When only "what each order cost" matters, workforce planning may stop being a schedule and become a cost curve.
By Dr. Chen, Supply Chain Researcher
