Research
“How safe am I at 100 mph in a car with 100,000 miles on it?”
In 2002 an Idaho State Police lieutenant wrote the fleet manager’s side of the story for the FBI — when to replace a patrol car, whether to lease it, and what happens when the budget says keep it another year.
The 1973 National Bureau of Standards study priced patrol cars from an economist’s desk: depreciation curves, discount factors, break-even models. It is rigorous and completely bloodless.
Thirty years later the FBI’s Law Enforcement Bulletin published the same subject written by somebody who had to live with the decision. Lieutenant Curtis W. Exley of the Idaho State Police, in Coeur d’Alene, laid out how a department actually decides when a police car is finished — and quoted a colleague making the argument that no spreadsheet contains.
Where the money goes
Exley opens with the ranking. Transportation equipment costs are the second greatest expenditure a law enforcement agency faces, behind only personnel salaries and benefits.
That refines the NBS claim rather than contradicting it. The 1973 study said patrol vehicles cost more than any other item of equipment. Exley is ranking against the whole budget. Both land in the same place: after the people, the cars are the biggest thing a police department spends money on.
How a police car actually gets bought
The mechanics are less formal than you would expect. An agency draws up a wish list of what it wants in a vehicle. That list becomes the specification. The specification goes out to area auto dealers — or in some cases corporate representatives — who calculate a price and bid.
Then Exley describes the instrument that turns that wish list into a decision, and it is the one this whole archive is built on. Performance results, he writes, are plugged into a unique competitive bidding formula: one percent of the lowest bid becomes the bid adjustment figure, and better performance across six weighted categories from top speed to fuel economy earns a dollar-and-cents advantage.
He is describing the Michigan programme — quoting a 1995 trade-press account of it — as the model other agencies aim at. Which is the clearest evidence in any of these documents of how far Michigan’s influence reached: an Idaho lieutenant explaining it to a national FBI readership as the way procurement ought to work.
A fourth founding date
Exley writes that Michigan has conducted these tests every year since 1978, citing a November 1995 article in Law Enforcement Technology.
That is now the third date in circulation. The National Institute of Justice says every fall since 1975. Michigan’s own books — the 1980 report, and the 1996, 1999 and 2001 editions counting their own anniversaries — point to 1976. Exley’s source says 1978.
1978 is explicable: the first published report covers the 1979 model year, tested in autumn 1978. Somebody reading from the shelf rather than the preface would reasonably conclude the programme began then. It is a reminder that the 1975 figure did not spread because anyone was careless — it spread because secondary sources cite each other, and the primary documents are harder to get.
One agency, one bid, many agencies
A detail here explains something about American police fleets that is otherwise puzzling. Smaller agencies can attach themselves to a larger agency’s bid, acquiring vehicles under its umbrella and sharing the discount that a bigger order earns.
The catch is that they must then accept the standards the larger agency endorsed. A small-town department’s patrol car is frequently not what that department would have chosen — it is what the state or the county chose, bought at the state or county’s price.
Lease or buy, and why nobody agrees
Exley reports that Michigan ran financial models through several versions and scenarios, and every time the models showed leasing saved the state money — largely because leasing deferred capital expenditure and flattened the budget, with consistent payments year after year rather than years where the agency has to ask for more.
And yet, he writes, most administrators responsible for police fleets still believe buying is the only sensible path.
His explanation of the disagreement is the sharpest analysis in the article. When fleet managers compare their own fleet against a lease, they typically leave out the cost of money, because it does not appear on their budget line. They account for depreciation and stop there. Lease costs include interest; ownership costs quietly do too, and are not counted. Compare the two and ownership looks cheaper than it is.
The argument between leasing and buying was not really about money. It was about which costs show up on the form.
When is a police car finished?
Exley polled agencies, and the spread is remarkable for a single question with a single right answer.
| Practice | Mileage |
|---|---|
| Shortest rotation reported | 45,000 miles |
| Longest rotation reported | 150,000 miles |
| Optimum rotation, per most sources | 50,000–70,000 |
| Mechanical repairs rise substantially after | 70,000–75,000 |
| Common practice: primary service to 50,000, then backup, low-mileage duty or auction | — |
Some agencies are bound by state law to withdraw a vehicle at a specific mileage. Others run to three times that. The same job, the same kind of car, and a three-fold difference in when it is considered used up.
The part that fails is not the engine
The failure analysis is the most useful passage for anyone maintaining one of these cars now.
Drivetrain components, Exley writes, usually incur only the minor problems a high-mileage vehicle will suffer, provided the prescribed maintenance intervals are followed. It is the ancillary parts that break. Suspension components wear out or weaken and break. Bushings deteriorate. Brake lines, fuel lines and wiring harnesses wear through at body attachment points or where they pass through frame members.
Interior components — seats, controls, upholstery — show wear thousands of miles before any mechanical component does.
And he is blunt about rust: road salt in winter states, and salt in the atmosphere in coastal ones, are the ideal recipe for a short vehicle life. Any restorer already knows which cars survived and which dissolved. This is the fleet manager’s version of the same knowledge.
The sentence that carries the article
Everything above is arithmetic. Then Exley quotes an officer, and the register changes completely:
“When I’m chasing someone at 100+ mph in a car with over 100,000 miles on it, how safe am I?”
The reasoning that follows is unanswerable. Even with the best of care and detailed safety inspections, potential problems can go undiscovered — and while many parts failures are merely irritating at low speed, at high speed they can be disastrous.
He notes something that makes it worse over time, too: as speed limits rose across the United States, patrol cars had to use higher and longer sustained speeds to overtake violators — putting additional stress on an already taxed brake system. The job got harder while the budgets got tighter.
And the public-facing version of the same problem: how safe will citizens feel when they see a ten-year-old patrol car? They may applaud the fiscal responsibility, and they may also ask why the cars are not safer.
Why it belongs here
Because it closes a circle. The NBS study worked out the mathematics of replacing a patrol car in 1973. Michigan built the instrument that made performance count against price. And here is a serving lieutenant in 2002, citing both, explaining to every police chief in America how to use them — and then pointing out that none of it answers the officer at a hundred miles an hour wondering what is about to break.
The cars in this archive were all bought under these rules, and sold under them too. Every survivor is a vehicle that reached somebody’s rotation number and was let go.