Should districts offer device insurance or protection plans?

The three models, and what each one actually buys you

Manufacturer accidental damage protection is bought per device at purchase, usually for a term matching the warranty, and typically covers a limited number of incidents per device per term. It is administratively the simplest option because it lives with the hardware and there is nobody to sell it to, but it is priced by the vendor for the average district, and it covers the vendor's repair, on the vendor's timeline. If you self-repair in house, you are paying for a service you partly duplicate.

Third-party family insurance is a product a company sells to your families, usually with a per-year premium and a deductible per claim. The district's role is to make it available and, generally, to be the entity that files and receives. The appeal is that the money and the risk sit outside the district; the drawback is that you have added a vendor to a family-facing process, and enrollment is voluntary, so the families most likely to opt out are frequently the ones least able to pay a repair bill.

District self-insurance is the model most districts converge on: a modest annual technology fee, waived for families who qualify, that funds an internal repair budget covering accidental damage. The district keeps the premium, keeps the repair work, and keeps the flexibility to decide what is covered. It is also the model that requires you to actually be able to do the math, because you are now the underwriter.

A fourth option is worth naming plainly: no insurance at all, with a published damage fee schedule and case-by-case waivers. For small fleets, or districts whose measured damage rate is low, this is a defensible answer and it avoids building an administrative apparatus that costs more than the damage it manages.

Running the break-even, with your own numbers

The core comparison is simple arithmetic and everyone skips it. On one side: the annual cost of coverage across the fleet, whether that is the per-device protection plan price or the premiums you forgo by not charging a fee. On the other: your actual annual out-of-pocket repair spend for the incidents that coverage would have absorbed. If your repair spend is consistently below the cost of the coverage, the coverage is not paying for itself, and that is true no matter how uncomfortable a bad year feels.

The inputs you need are damage rate per hundred devices per year, cost per repair by symptom, and the mix of symptoms. A district that mostly sees cracked screens has a very different exposure from one that mostly sees liquid damage and dead boards, because the first is a stocked part and forty minutes of labor and the second is often a total loss. Both districts might have identical incident counts.

Be honest about the labor side. If you self-repair, the parts cost understates the true cost, and if you are choosing between vendor coverage and in-house work, the technician hours are the whole point of the comparison. Districts that track parts cost and technician time per repair can make this call with evidence; districts that track neither have a belief.

Do not average across the fleet if the fleet is not uniform. Damage rates differ sharply by grade band and often by building, and the honest version of the analysis is per-cohort. That matters because it may lead you to a per-cohort answer — coverage for take-home high school devices, none for elementary carts that never leave the building — which is a better decision than one number applied to everybody.

Finally, run the analysis against a distribution and not just a mean. Self-insurance fails when a bad year arrives before the reserve is built. If a single severe year would exceed what you can absorb, that is an argument for external coverage even when the average says otherwise.

Equity, waivers, and the constraints you cannot design around

A device fee that gates access to a device is a legal and ethical problem, and in many states it is squarely a legal one. Public education is free, and if a Chromebook is the instrument of instruction, conditioning access on payment tends to run into the same constitutional and statutory limits as charging for textbooks. Check your state's rules and get your attorney's read before you publish a fee schedule, not after.

Whatever you charge must therefore be genuinely optional in effect: a student whose family does not pay still gets a device and still participates. That is the design constraint, and it rules out the simple version of insurance where non-payers are excluded from coverage. What most districts do instead is charge a modest fee, waive it automatically for families qualifying for free and reduced lunch, and extend coverage to everyone regardless of payment — accepting that the fee is a partial offset rather than a full one.

The collection reality reinforces this. Assessed is not collected, and the gap is often large. Build your model on expected collections rather than on assessed amounts, and track the two separately so you know which one you are looking at. A protection scheme that only works at high collection rates does not work.

Be careful with third-party products specifically. If a vendor is selling insurance to your families using your name, your rostering data, or your communication channels, you have questions to answer about data sharing, about what you are implicitly endorsing, and about what happens when a claim is denied and the family calls the school. Some districts are comfortable here and some are not; the point is to decide deliberately.

Designing the coverage so it changes behavior the way you want

Any coverage design sends a signal, and the signal matters more than the dollars. Full coverage with no family responsibility removes the incentive to be careful, which measurably increases damage rates in some districts. Zero coverage with full replacement cost billing produces hidden damage — devices that come back broken with nobody admitting when, students who stop reporting problems, and a repair queue that fills in June instead of continuously.

The common middle is a small per-incident amount the family owes regardless of coverage, with the rest absorbed. It preserves the incentive without creating a bill families cannot pay, and it keeps reporting honest, which is worth more than the money.

Graduate the response for repeat incidents rather than the first. A first accidental break is a normal thing that happens to a fourteen-year-old; a fourth is a pattern, and a policy that treats them identically is either too harsh at the start or too permissive at the end. Escalating family responsibility across incidents within a year, with a documented appeal, is defensible and understood.

Draw the line between accidental damage and negligence or intent in writing, and accept that the line will be argued. A cracked screen from a dropped backpack, a device left on a bus, and a device with a key deliberately pried off are three different situations. Give building administrators the discretion to make the call and give families a route to appeal it, because the alternative is that the technology department becomes the disciplinary authority, which it should not be.

Whatever you decide, put it in the device agreement families sign at issue, in plain language, with the actual amounts. The number of disputes is roughly proportional to how surprising the bill is.

What to track so next year's decision is better than this year's

This decision should be revisited annually with evidence, which means instrumenting it now. The minimum set: incidents per hundred devices per year, split by building and grade band; cost per incident including parts and labor; the symptom mix; the amount assessed versus the amount collected; and, if you carry vendor coverage, how many claims you filed and what they returned.

The claim-rate number is the one that most often changes minds. Districts frequently buy accidental damage protection and then, at the counter under time pressure, repair in house without checking coverage — paying for both. If you cannot say how many covered claims you actually filed last year, you do not yet know whether the coverage is working, only that you bought it.

Track repeat incidents per device and per student separately. Per device tells you whether a model is fragile; per student tells you whether a handling or supervision problem exists that no insurance product will fix. Those are different interventions and the aggregate damage rate hides both.

Measure the administrative cost too, at least roughly. Time spent enrolling families, filing claims, chasing payments, and answering questions is real cost, and it is the cost most often left out of the comparison. A self-insurance scheme that saves money on paper and consumes a half-time position has not saved money.

After one full year with those numbers you can answer the question directly rather than by analogy to what the district next door does — which is how most of these decisions currently get made, and why so many districts are carrying coverage that their own data would not justify.