What an Empty Position Actually Costs You

You will be asked what the vacancy costs. Before you answer, understand what the person asking already believes.

• On a budget line, an empty position looks like money saved. That is the argument you are up against.
• There is no national cost-per-officer figure. The ones circulating mostly cite sources that do not contain them.
• The cost is real, but you have to build it from your own documents. Here is the worksheet.

Your finance office already has a number

Some state budget offices formally book vacancies as savings. One state's legislative fiscal office calls it a turnover savings factor — it averages two years of appropriated-versus-actual salary spending, derives a historical turnover rate for each agency, and applies that rate when personnel budgets are calculated — so the money is netted out before the year begins.¹

That is not sloppy accounting. It is correct, from where they sit. Salary you do not pay is money you do not spend.

Which means the case you bring cannot be that vacancies are bad. Everyone agrees vacancies are bad. The case has to be that the empty position costs more than the salary it saves — in dollars, from documents your city already publishes.

There are four of those costs. Only one of them ever shows up on the staffing line.

What you already spent on the person who left

One Pacific Northwest city published the full arithmetic in 2026. Bringing a single officer through probation costs about $345,000.²

Look at what dominates. Salary and benefits during training are 87% of it. The academy tuition everyone argues about is 6%.

That matters, because it is the line most often left out. Ask what it costs to train an officer and you get an academy invoice — one state academy publishes its full basic course cost at $14,400 for 620 hours.³ True, and about 4% of the real number. You are paying someone a full salary for close to two years before they work a shift alone.

The same report notes about 25% of trainees separate before completing probation, and states the consequence directly: the bureau “must invest in approximately four trainees to yield three deployable officers.” At that ratio every officer who finishes has absorbed the cost of one and a third starts — nearer $460,000. The ratio is the city’s; the multiplication is ours.

What the empty shift costs while you look

This is the best-documented cost in policing, because city auditors keep measuring it.

The auditor’s mechanism is the plain one: “When an officer takes leave, another officer must work overtime to meet minimum staffing requirements, increasing Police’s overtime expenses.”⁴

Another city auditor tracked the two together directly. Officer vacancy rates went from 7% to 10%, and overtime hours worked rose 54%. That department spent more than $50 million on overtime in one year, about 9% of the department’s total expenditures, with 156,540 hours booked specifically as patrol backfill.⁵

Now do the step none of the audits does for you. A full-time position is about 2,080 hours a year. 156,540 hours of patrol backfill is roughly 75 officers’ worth of work, bought one overtime shift at a time, at premium rates. That division is ours, not the auditor’s — but it is arithmetic anyone can check, and it is the most persuasive line you can put in front of a city administrator.

What it costs when the community is short

The one peer-reviewed dollar figure we could find measures something different from everything above.

Researchers studying one western agency of roughly 600 officers used RAND’s cost-of-crime calculator to price the resignations it experienced. Losing 33 officers projected to $11.64 million in additional crime costs over a year. The study separately values each individual officer at $357,533.⁶ Those are two independent outputs of the calculator, not one divided by the other — do not present them as if they were.

The number nobody publishes

Before you cite a national figure, know that there isn’t one.

The federal government’s most recent review of police recruitment and retention, published February 2026, contains no per-officer cost figure for recruiting, hiring, academy, field training, turnover or vacancy.⁷

Meanwhile the ranges in circulation — $100,000, $111,000 to $190,000, $20,000 to $100,000 — are published either with no citation at all or credited to sources that do not contain them. We chased them. One national policy institute states the $20,000 to $100,000 range with nothing attached to it.⁸

The closest thing to an official number is a ceiling rather than a cost: the federal hiring grant program caps its share at $125,000 per officer position over three years, covering up to 75% of entry-level salary and benefits.⁹

Bring a sourceless national range to a budget meeting and you will lose on the first question. Bring your own agency’s arithmetic and there is nothing to attack.

Build your own number

Every input below already exists in a document your agency or city has. None of it requires a study.

1. Pull one year of backfill overtime. Hours, not dollars, then apply your own average overtime rate. Hours are harder to argue with.
2. Divide those hours by 2,080. That is how many officers you are already paying for and do not have. This single number does more work than everything else on the page.
3. Build the sunk cost from payroll, not from memory. Recruit salary and benefits for the full pre-solo period. That is your biggest line and the one people forget.
4. Divide by your completion rate. If three in four finish probation, every finisher carries the cost of one and a third starts.
5. Multiply the coverage cost by your real cycle time. A vacancy does not cost a year. It costs however long your process actually takes.
6. Show the sources next to every line. A number a city analyst can trace is a number they cannot dismiss.

The argument you will get back

Expect this one, because it is not stupid: overtime is cheaper than headcount. An officer paid time-and-a-half for 200 extra hours costs the city several thousand dollars. A new officer costs six figures a year, every year, for a career, plus pension.

On a one-year spreadsheet that argument wins. Three things beat it, and all three are yours to produce.

• The hours are not marginal. When one audited department books 156,540 hours of patrol backfill in a year, that is not a few extra shifts. Convert it to officer-equivalents and say the number out loud.
• Overtime does not fall when vacancies do. One auditor put it plainly: “while the number of vacancies has decreased, overtime has not declined at the same rate.”¹⁰ Our read is that chronic backfill goes structural once the schedule is built around it. Either way, raise this before someone raises it at you: overtime bought as a stopgap does not get given back.
• The cost is not only the money. Backfill is worked by the officers you already have, which is a retention cost you pay later at full price.