Two cost lines erode care home margins faster than almost anything else: agency staff spend and occupancy voids. Neither is fully controllable, but both are manageable once you have a finance model behind them rather than a gut feel. This guide builds that model, step by step, from verified figures.
For the worked numbers behind staffing costs, use the true cost of a care hour calculator and the staffing margin calculator alongside this guide.
Why agency staff cost so much more than your own
Agency cover carries the agency's commercial margin on top of every underlying employment cost the worker attracts. That margin is the core reason agency spend erodes profit so quickly. A substantive worker costs pay plus employer NIC at 15% above the secondary threshold of £5,000, plus holiday accrual and any mileage. An agency worker costs all of those things, packaged into the invoice rate, plus the agency's own margin layered on top.
The secondary effect is scheduling disruption. Agency workers are unfamiliar with residents, slower to integrate into rotas, and often unavailable at the exact shift length needed, so blocks of agency time frequently exceed the actual gap they fill. The invoice grows beyond the true hours needed.
The case for reducing agency reliance is therefore not just the visible premium on the invoice rate. It is the compound effect of margin, inefficiency and the ratchet dynamic: as agency use rises, recruitment of substantive staff slows because the immediate gap is always filled, which sustains agency dependency.
The true cost of a care hour, built up
The table below builds the loaded cost per hour for a substantive care worker under 2026-27 rates, then contrasts it with an illustrative agency invoice rate to show where the gap sits. The agency rate is illustrative only; actual rates vary by region, shift type and provider relationship.
| Cost element | Basis | Per-hour amount (illustrative) |
|---|---|---|
| Base pay (NLW floor) | £12.71 per hour from 1 April 2026 (21 and over) | £12.71 |
| Employer NIC | 15% on earnings above £5,000 per year (£96 per week); modelled per head at full-time hours | approx £1.73 |
| Holiday accrual | 12.07% of hours worked for irregular-hours staff | approx £1.53 |
| Mileage (where applicable) | 55p per mile from 6 April 2026, first 10,000 miles; spread across chargeable hours | varies by role |
| Loaded substantive cost | approx £15.97+ per hour | |
| Agency invoice rate (illustrative) | Includes the agency's margin on top of the above; actual rate varies | your contract rate |
| Agency premium | The gap between the two lines is the cost of reliance | track per period |
The employer NIC figure above uses a per-head model rather than a blanket percentage, because individual workers crossing the £96 weekly secondary threshold at different hours trigger NIC at different effective rates. The true cost of a care hour calculator runs the per-head model for your actual workforce.
Note on the NIC build: the secondary threshold is £5,000 per year (£96 per week), with the 15% rate on earnings above that level. A part-time worker earning below £96 per week generates no employer NIC at all, which is why staffing mix (full-time versus part-time) materially affects the aggregate NIC cost.
Controlling agency spend without breaching NMW or misclassifying
The most common lever is building a bank of zero-hours workers, typically former employees or known candidates, who can cover sickness and unplanned leave without triggering an agency invoice. The saving is the agency's margin: the underlying worker cost is similar. But bank staff carry the same legal obligations as any other worker.
Travel time between shifts is working time for NMW purposes. A bank worker driving from one care home site to another between back-to-back shifts must be paid for that travel at or above the applicable NMW rate. Paying only for face-to-face contact time is the most common NMW underpayment pattern in the sector. Gov.uk confirms that work-related travel is working time; commuting to the first shift and from the last shift home is excluded.
Sleep-in shifts: only time awake for the purposes of working counts for NMW. The Supreme Court settled this in Royal Mencap Society v Tomlinson-Blake [2021]. Workers on sleep-in shifts who are permitted to sleep are not entitled to NMW for the sleeping period. The awake-for-the-purposes-of-working test applies; the position changes if sleeping facilities are not provided or the worker must remain active throughout.
Bank and zero-hours staff cannot be labelled self-employed to cut costs. Workers on rotas who cannot send substitutes, are told when and where to work, and are subject to the provider's direction will be employees or workers in substance, regardless of the label on their contract. HMRC audit activity in adult social care is focused on exactly this pattern. Backdated PAYE, NIC and NMW liability can extend years into the past.
Holiday pay must be included. Irregular-hours workers accrue holiday at 12.07% of hours worked in each pay period. Errors in holiday pay calculation, particularly failing to include average travel payments in the reference period, are a frequent Employment Tribunal category in care.
See also: Sleep-in pay and travel time NMW: a care operator guide.
The Employment Allowance offset on your substantive wage bill
Most care operators are eligible to claim the Employment Allowance, which reduces employer NIC by up to £10,500 per tax year. For a smaller home or domiciliary agency whose total employer NIC liability falls below £10,500, the allowance eliminates the bill entirely.
This matters for the agency-versus-substantive comparison. When you recruit a substantive worker to replace agency cover, the employer NIC on that worker's pay is partly or fully offset by the allowance for eligible operators. The agency, billing you at an all-in rate, has already priced their own NIC cost into the invoice with no offset available to you. The Employment Allowance therefore tilts the comparison further toward substantive employment for operators who have headroom left in their allowance.
Larger groups with associated entities see the allowance spread across the group rather than stacked per entity, so the calculation needs to be done at group level. The care homes finance hub sets out how this interacts with employer NIC planning for multi-site operators.
Occupancy voids: the second margin lever
An empty bed earns nothing but still carries most of its fixed cost. That asymmetry is why occupancy voids fall almost entirely to the bottom line rather than reducing costs proportionately. The week a bed is empty, the rent or mortgage, core staffing, insurance and utilities all continue. Only a small slice of direct variable cost disappears.
The correct way to account for occupancy is on an available-bed-day basis: divide occupied bed-days by total available bed-days in the period. A bed that is physically present but temporarily out of service for refurbishment is not an available bed; a bed that is simply unfilled is an available, void bed.
Each void day should carry an explicit cost line in management accounts: the fixed overhead attributable to that bed, net of any variable saving. Burying void impact in an overall margin shortfall obscures the true driver and makes it impossible to model the revenue value of each additional filled bed.
| Item | Per bed-week (illustrative) | Notes |
|---|---|---|
| Weekly fee income (occupied) | your actual fee rate | Varies by fee payer: self-funder, LA, FNC |
| Variable cost per occupied bed | food, consumables, direct care hours above fixed rota | The only costs that genuinely disappear when a bed is void |
| Contribution per occupied bed-week | fee income minus variable cost | This is what a void day costs you in lost contribution |
| Fixed cost per available bed-week | total fixed costs divided by bed capacity | Carries whether the bed is occupied or not |
For the interaction between your fee mix (self-funders, local authority and FNC income) and occupancy accounting, see the FNC, CHC and LA fee-mix accounting guide.
Break-even occupancy, worked
Break-even occupancy is the occupancy percentage at which contribution from occupied beds exactly covers fixed costs. Below it the home makes a loss; above it, each additional occupied bed-week drops almost entirely to profit.
The model is straightforward. You need three numbers: total fixed costs per week, contribution per occupied bed-week, and total bed capacity.
- Divide weekly fixed costs by contribution per occupied bed-week to get the number of beds that need to be occupied to cover fixed costs.
- Divide that number by total bed capacity and express as a percentage. That is your break-even occupancy rate.
Worked illustration (hypothetical inputs):
| Input | Illustrative figure |
|---|---|
| Total bed capacity | 40 beds |
| Weekly fixed costs | £28,000 |
| Weekly fee per occupied bed (blended average) | £1,100 |
| Variable cost per occupied bed-week | £150 |
| Contribution per occupied bed-week | £950 (£1,100 minus £150) |
| Beds needed to cover fixed costs | 29.5 (£28,000 / £950) |
| Break-even occupancy | 73.7% (29.5 / 40) |
In this illustration, every bed-week above 29.5 occupied beds contributes £950 to profit. Every void week below that line costs the home £950 in lost contribution on top of the fixed cost that runs regardless. The model makes explicit why a drop from 90% to 80% occupancy is not a 10-point problem but a direct reduction in the number of contribution-generating beds.
Your actual figures will differ. The staffing margin calculator lets you run the model with your own fee rates, fixed cost base and bed count.
Reading agency spend and occupancy together
Agency spend and occupancy voids compound each other. When occupancy drops, income falls but fixed staffing must broadly continue, creating pressure to reduce bank and substantive shifts, which in turn increases agency reliance for the remaining occupied residents who still need care. The margin erosion is therefore faster than either driver would produce alone.
A simple weekly dashboard with three lines surfaces this dynamic:
- Agency-spend ratio: agency invoice total as a proportion of total staffing spend. Track week on week; a rising ratio signals structural reliance, not just temporary cover.
- Occupancy percentage: occupied bed-days divided by available bed-days, for the same period.
- Contribution per occupied bed-week: fee income minus variable cost, calculated on the actual beds filled. This number falls when the fee mix shifts toward lower-rate LA placements or when void days increase.
What "good" looks like will vary by home size, fee mix and cost base. There is no universal benchmark that applies across the sector. The value of the dashboard is the trend, not a comparison to an external number. A rising agency ratio combined with falling occupancy and falling contribution per bed is the pattern that needs urgent attention.
Skills for Care publishes annual workforce and sector data, including turnover and vacancy rates across adult social care in England, at their state of the adult social care sector report. That is the authoritative source for any sector-level reference point you want to compare against.
Getting the numbers under control
The starting point is understanding your current position with precision: what does a care hour actually cost by worker type, what is your real occupancy on an available-bed-day basis, and what is the break-even occupancy for your cost base?
The true cost of a care hour calculator builds the loaded cost from your actual pay rates and hours. The staffing margin calculator models agency-versus-substantive comparisons and break-even occupancy with your own inputs.
The care homes finance hub covers how these cost lines interact with your fee negotiation position, your CQC financial viability requirements and your broader care business structure. The care provider business index tracks how many care companies are forming and dissolving each quarter, for market context.
If you want a finance review that covers staffing costs, occupancy modelling and the tax position for your home specifically, the care homes hub sets out what that engagement covers and how to get started.