TL;DR
- Depreciation sits under Section 34(F) of the National Internal Revenue Code. That part is settled.
- What is not settled is the number. No binding BIR useful life figure for laptops could be confirmed, despite plenty of articles stating one with confidence.
- Practitioners report three to five years for IT equipment. A commonly cited 20 per cent rate traces to a 1940 regulation we could not verify as still operative.
- Changing a useful life or a method later is reported to need BIR approval, so the first choice matters more than the second.
- The expensive mistake is not the number. It is the gap between the book life and your actual refresh cycle.
- Nothing here is tax advice. Confirm the current treatment with a Philippine accountant before you rely on it.
Somebody in finance asks how long a laptop lasts. It sounds like a question with an answer. It is not.
Search for it and you will find a great many confident numbers. Three years. Five years. Twenty per cent a year. Each is presented as though it came from a table somebody could point at. Go looking for that table and it becomes harder to find than the confidence suggests. We spent a while looking, and what is actually published is thinner than almost every article on the subject implies.
Which matters, because this is one of the most searched equipment questions in the Philippines. Finance teams and IT teams both need an answer, they need the same answer, and they frequently arrive at different ones because they are solving different problems. Finance wants a defensible treatment. IT wants to know when it can replace the machine. Those two numbers drift apart, quietly, for years.
Best tools for Device Management
So this page does two things. It says plainly what is published and what is not, with the instrument named. And it covers the operational question underneath, which is the one that actually costs money.
A boundary first. Nothing below is tax advice and none of it states what is required of you. Treatment depends on your circumstances, and the right answer comes from a Philippine accountant who knows them. This page describes what the agencies publish so you know what to ask about.
What is actually published, and what is not
Start with the part that is clear.
Depreciation of business assets in the Philippines sits under the National Internal Revenue Code, Section 34(F), as amended. Claims are expected to be supported by purchase documentation and a depreciation schedule. And changing a useful life or a method once adopted is reported to require approval from the Bureau of Internal Revenue rather than being an annual choice.
Now the part that is not clear, which is the number itself.
We could not confirm a binding BIR useful life figure for laptops from a primary source. What exists instead is a body of professional practice. Accounting firms and consultancies report three to five years as the range commonly accepted for IT equipment. A frequently cited 20 per cent annual rate, which implies a five year life, is attributed to Revenue Regulations No. 2-40. That regulation dates from 1940, and we could not verify that it remains the operative authority on this point.
Sources also disagree with each other on method. Some state that the straight line method is required under Section 34(F) as amended by the CREATE Act. Others say other methods are permitted where prescribed, and that no specific rules have been issued on useful life or method at all. Both positions are published by firms that do this work professionally. We are not in a position to resolve the disagreement and will not pretend otherwise.
So the honest summary is this. The statute is clear. The documentation expectation is clear. The specific number for a laptop is professional judgement rather than a published figure, and anybody telling you it is settled has not gone looking.
Why the confident answers in circulation are worth less than they look
This is worth a short section, because the pattern repeats across every market and it is useful to recognise.
An article states a figure. A second article cites the first. A third cites the second. By the fourth, the number has the weight of consensus and no traceable origin. Follow the chain back far enough on laptop depreciation philippines questions and it usually ends at another blog post rather than at a regulation.
The tell is specificity without citation. A page that says “the BIR allows five years” and does not name the instrument has not read one. A page that says “Section 34(F) governs depreciation, and practitioners commonly apply three to five years for IT equipment” is telling you both what it knows and how firmly.
But there is a more useful reason to care than accuracy for its own sake. If you adopt a number because an article said so, and somebody later asks you to justify the treatment, “it is the standard” is not an answer. A documented basis is. So the work is not finding the right number. It is being able to say why you chose the one you did.
Which, conveniently, is also the thing your accountant will ask you first.
There is a wider version of this worth carrying into any question about laptop depreciation philippines rules, or any other jurisdiction for that matter. The figure that survives scrutiny is rarely the one that was most common online. It is the one somebody chose for a stated reason, applied consistently, and can still explain three years later. That standard is lower than it sounds and almost nobody meets it, because the number gets picked in five minutes by whoever needed the schedule to balance.
Useful life and refresh cycle are different numbers, and the gap costs money
Here is the part that matters operationally, and it gets almost no attention compared with the tax question.
Useful life
is a book figure. It decides how the cost is spread across years in your accounts. Refresh cycle is an operational figure. It decides when somebody actually gets a new machine. They are set by different people, for different reasons, and nobody checks whether they match.
When the book life is shorter than the refresh cycle, you end up running a fleet of fully depreciated machines. They carry no book value, so they appear on no financial report, and the organisation loses visibility of a real asset base. People are working on equipment nobody is tracking the age of, because the only system that was tracking it stopped caring at year three.
When the book life is longer than the refresh cycle, every refresh generates a disposal loss. The machine leaves at three years carrying two years of unamortised cost. Do that across forty devices and somebody in finance will eventually ask why every hardware cycle produces a write-down, and the honest answer is that the assumption was wrong rather than that anything went wrong.
Neither outcome is a disaster. Both are avoidable by having one conversation. And the conversation is short: what is our actual replacement cycle, and does the book treatment reflect it.
Pick the refresh cycle first, then the book life
The instinct is to take the tax answer and work backwards. That gets it the wrong way round, because the refresh cycle is the thing you control and the thing that affects whether people can work.
What should drive it is specific. How heavy is the workload, since a machine running development tooling ages faster in practice than one running a browser and a spreadsheet. What is the support cost curve, because repair frequency rises and somebody is paying for it in time if not in cash. What is the security position, since older hardware eventually stops receiving the operating system versions you need. And what is the residual value, which falls fastest in the first two years and flattens after.
Most teams landing on three years are landing there for good reasons. Warranties commonly run three years. Residual value is still real at that point. The security position is comfortable. And the machine has not yet entered the phase where repairs get frequent.
Teams landing on four or five are usually either running light workloads, or managing cash carefully, or both. That is a legitimate choice. It needs a larger spares buffer, because failure rates rise, and it needs somebody watching the operating system support position rather than assuming it holds.
Once the operational cycle is settled, take it to your accountant and ask whether the book treatment can reflect it. Often it can. Where it cannot, at least the mismatch is now known and deliberate instead of accidental.
What goes into the cost, which is messier for an imported machine
Before anything can be depreciated, somebody decides what the asset cost. For a laptop bought locally that is close to trivial. For an imported one it is not, and this is where Philippine specifics bite.
A machine shipped into the country arrives carrying more than its invoice price. There is freight. There is insurance. There is 12 per cent VAT assessed on the customs value, which includes that freight. There are customs processing and documentary charges. And there may be broker fees. Which of those form part of the asset cost, and which are period expenses, is a question with a real answer and it is not one this page should give you.
What this page can say is that the question exists, that it is frequently answered by default rather than by decision, and that the default is usually whatever the person entering the invoice did first. The amounts are not trivial. On a fleet purchase the non-invoice elements can reach a meaningful share of the total, and treating them one way rather than another changes both the depreciation base and the current year position.
So the practical step is to keep the components separate in your records rather than collapsing them into one figure at entry. Invoice price, freight, insurance, tax, fees, each identifiable. Then your accountant can apply whatever treatment is correct. Collapse them at entry and reconstructing the split later means going back through courier paperwork, which is exactly as pleasant as it sounds.
And this is one more argument for buying locally where the specification allows. A domestic purchase has one number on one invoice, and the whole question disappears.
Capitalise or expense, and the threshold nobody wrote down
Not every piece of equipment goes on a depreciation schedule. Somewhere there is a line below which things are simply expensed, and in most organisations that line exists in somebody’s head rather than in a policy.
That is fine until it is not. The problem is consistency. The same class of item treated one way in March and another way in September produces a schedule nobody can explain, and the explanation is usually that two different people made reasonable decisions without a shared rule.
Laptops almost always sit above any sensible threshold and get capitalised. The genuine ambiguity is in the things around them. A monitor. A docking station. A headset. A memory upgrade fitted a year after purchase, which raises a separate question about whether it extends the asset or is a running cost. An extended warranty bought with the machine.
Each has an answer. None of the answers should come from here, because the threshold and the treatment are things to agree with your accountant and then write down. What matters operationally is that the rule exists, is written, and is applied by whoever enters invoices rather than decided fresh each time.
One practical note. Where peripherals are expensed rather than capitalised, they drop off the financial record entirely, which means your asset register is the only place they exist. That makes the register more important rather than less, because it becomes the sole account of a category of equipment that still has to be recovered when somebody leaves.
What finance needs from IT, and never gets
A depreciation schedule needs four things per machine. Finance can produce none of them without IT, and IT rarely knows they are wanted.
The purchase date, which is when depreciation starts, and which is frequently approximated from an expense claim rather than recorded. The cost, which should include what was capitalised alongside the machine. The serial number, because it is the only identifier that survives a rebuild and a change of holder, and because without it two identical machines are indistinguishable on a schedule. And the disposal date and proceeds, which close the entry.
That fourth one is where schedules go wrong most often. A machine leaves the business and nobody tells finance. The asset stays on the schedule, depreciating quietly, long after it has physically gone. Multiply by a few years of turnover and the fixed asset register contains machines that do not exist, which is a problem somebody will eventually have to unwind by hand.
The fix is unglamorous and it works. Whoever closes the asset record in IT also notifies finance, in the same step, with the serial number and the date. One step. It fails only when it is two separate processes owned by two separate people.
Fully depreciated and still working
A machine that has reached the end of its book life has not stopped being useful. Treating those two facts as the same thing is a common and expensive error in both directions.
Some organisations replace on the book schedule, because the asset is “done”. That throws away working equipment on an accounting trigger rather than an operational one. A three year old machine doing light work may have two good years left, and the redeployment value is real.
Others keep machines indefinitely because they cost nothing on paper. Which is true and misleading. A fully depreciated laptop still has a support cost, still has a failure rate that is climbing, and still has an operating system that will eventually fall out of support. It appears free precisely because the system that was measuring it has stopped.
So the useful discipline is to track age separately from book value. Your asset register already holds the purchase date. A quarterly look at the age distribution tells you something the financial report cannot, which is how much of your fleet is in the phase where things start going wrong.
And that distribution is usually worse than people expect the first time somebody plots it. Fleets age invisibly, one machine at a time, and nobody notices until several fail in the same quarter.
Disposal, residual value and the number that surprises people
The end of the schedule is the disposal entry, and it is where the book figure meets the market.
A three year old business laptop retains real value. Not much, and it varies enormously by model and condition, but it is not zero, and a machine written down to zero that sells for something produces a gain somebody has to account for. Equally, a machine still carrying book value that sells for less produces a loss. Neither is a problem. Both need recording.
What catches people out is the difference between the three disposal routes. Resale produces proceeds and needs the data dealt with first. Redeployment produces no proceeds and keeps the asset in service, which is usually the best outcome and the one most often skipped because nobody holds the machine. Disposal through a recycler frequently produces nothing or costs money, and it is the right answer for genuinely end of life hardware.
Where a device cannot be recovered at all, which happens, the entry still has to close. A write-off with a stated approval threshold by value keeps the register accurate. Without one, unrecoverable machines sit on the schedule indefinitely because nobody has the authority to remove them, and the register slowly becomes fiction.
None of this is complicated. It simply needs somebody to own the closing entry, which is the same person who should own the recovery.
Where the asset sits when you have no Philippine entity
A question that arrives constantly and gets very little written about it. You have people working in Manila. You have no company there. The laptops are in the Philippines. Whose balance sheet are they on, and where are they depreciated?
There is no single answer, and the answer depends on structure rather than on geography. A machine owned by a parent company abroad and used by somebody employed through a local provider is in a different position from one bought by a Philippine subsidiary, and both are different again from one supplied by an outsourcing provider as part of a service.
What this page will say is which facts determine it, so you can bring the right ones to the conversation. Who bought the device and from where. Who legally owns it. Which entity, if any, imported it and appears on the customs documentation. Whether there is a local entity at all. And whether the device is supplied to the worker by you or by a third party.
That customs point is the one people miss. Where a machine was imported, the import documentation names an importer, and that record exists independently of whatever anybody later decides about ownership. If the entity on the customs paperwork is not the entity carrying the asset, somebody will eventually ask why, and reconstructing the reasoning years later is hard.
So the practical advice is narrow. Decide the ownership position before the first import rather than after, keep the customs documentation with the asset record, and put the question to an accountant who can see your actual structure. Anybody answering it without knowing whether you have a local entity is guessing.
And this is a further reason the local purchase route is cleaner for small numbers. No import, no importer of record, no documentation trail to reconcile against a balance sheet in another country.
Damage, loss and the mid-life write-down
Schedules assume a machine runs its course. Plenty do not, and the exceptions need a route through the process or they become permanent anomalies.
Three cases come up. A machine is damaged beyond economic repair partway through its life, which in the Philippines includes a real share of liquid damage and transit damage. A machine is lost or stolen. And a machine is never recovered from somebody who left, which is the most common of the three and the one least likely to be reported promptly.
In each case the asset still carries book value and is no longer an asset. How that is treated, whether as a write-down, a disposal or something else, and what documentation supports it, is a question for your accountant. What IT controls is whether finance finds out at all, and when.
The practical mechanism is a single trigger. Whoever marks a device as written off in the asset register notifies finance in the same step, with the serial number, the date and the reason. Not a monthly report. Not a conversation at year end. The same step.
Insurance deserves a mention here too, because it changes the arithmetic. Where devices are covered, a loss produces a claim and proceeds, which affects the entry. Where they are not, the whole remaining value goes. Knowing which applies before something happens is better than finding out during.
And the underlying point is the same as everywhere else on this page. The accounting is not hard. Being told is the hard part.
Does this change the buy or lease decision
Sometimes, and less often than vendors suggest.
Buying puts the asset on your books and spreads the cost across the useful life. Leasing or a device-as-a-service arrangement generally converts it to an operating expense, which changes how it appears and when. Which is better depends on your cash position, your tax position and how you are measured, and that is a genuine finance question rather than an IT one.
What is worth knowing from the equipment side is that the arguments made on tax grounds are usually smaller than the arguments made on operational grounds. A subscription that includes replacement, support and recovery across the Philippines may be worth more for those reasons than for how the cost is classified.
And the comparison is frequently run badly. A purchase price is compared with a monthly fee, and the purchase side omits the things the subscription includes: spares, replacement logistics, recovery, wiping, and somebody’s time. Add those and the gap narrows considerably. Omit them and buying always looks cheaper, which is why it usually does in the first version of the spreadsheet.
Run it properly over the full cycle, including the exit, and take the structure question to your accountant rather than to a vendor comparison page.
The decision table
| Situation | Scale | Setup | Primary Pain | Recommended Starting Point |
|---|---|---|---|---|
| No schedule exists | Under 20 | Expense claims only | Nothing is capitalised | Build the four-field list, then ask an accountant |
| Schedule exists, IT does not feed it | 20 to 100 | Two separate processes | Disposed machines still on the books | One step: whoever closes the asset record tells finance |
| Book life three, refresh at four plus | Any | Mismatched | Fleet ageing invisibly | Track age separately from book value, quarterly |
| Book life five, refresh at three | Any | Mismatched the other way | Disposal loss every cycle | Align the book treatment with the real cycle |
| Machines unrecoverable | Any | No write-off path | Register carries fiction | Approval threshold by device value |
| Weighing buy against subscription | 30 plus | Comparing price to fee | Comparison omits the hidden costs | Cost the full cycle including spares and exit |
The year end nobody plans for
Once a year somebody asks for a fixed asset listing, and the state of it is decided entirely by what happened in the eleven months before.
The pattern is predictable. Finance requests the schedule in the last week of the period. IT produces a list from whatever system it keeps. The two do not match. Somebody spends several days reconciling them, mostly by asking people what happened to specific machines, and the reconciliation produces a set of adjustments that look like errors but are actually a year of unreported changes arriving at once.
And then the same thing happens the following year, because the reconciliation fixed the numbers without fixing the process that produced them.
Breaking that cycle needs one change, and it is not software. It is that disposals and write-offs are reported when they happen rather than discovered at year end. Everything else follows. A register updated continuously needs a sample check, not a reconstruction.
There is a cheap intermediate step for teams not ready to change the process. A quarterly five-minute note from IT to finance listing only what changed: devices added, devices disposed, devices written off, by serial number. Four short notes a year replace one long week, and they catch the discrepancies while people still remember the circumstances.
Because that is the real cost of annual reconciliation. Not the days spent, but that the information needed to resolve a discrepancy decays. Asking in March what happened to a laptop in February is easy. Asking in December is archaeology.
The reconciliation almost nobody runs
One check tells you whether any of this is real, and it takes an afternoon.
Take the fixed asset register. Pick twenty laptop entries at random. For each one, confirm three things: that the machine physically exists, that the person named as holder still works for you, and that the serial number on the schedule matches the serial number on the device.
Most organisations running this for the first time find somewhere between two and six discrepancies in twenty. Machines disposed of but still listed. People who left last year. Serial numbers that were transcribed wrong at entry. None of it is dramatic and all of it compounds.
The point is not the twenty. It is the rate. Three wrong in twenty across a fleet of two hundred implies about thirty bad records, which is the difference between a register you can rely on and one you cannot. And the only way to know the rate is to sample it.
Run it quarterly on a fresh sample. It is cheaper than an annual full audit and it catches drift while the drift is still small, which is the entire point.
One refinement worth adding after the first run. Record not just how many were wrong but what kind of wrong, because the categories point at different fixes. Stale holders mean the leaver process is not feeding the register. Missing disposals mean the closing step is not happening. Wrong serials mean data entry, which is the easiest to fix and the least damaging. Three discrepancies of the same kind is a process problem. Three of different kinds is just entropy, and entropy is what the quarterly sample is for.
What to take to your accountant
Five questions, which turn this from an article into a decision.
What useful life should we apply to business laptops, and what is the basis for it. Which method should we use, and is there any constraint we should know about. What documentation do you need from us per device, and in what form. What happens if we want to change the treatment later. And how should disposals and unrecoverable devices be handled, including write-offs.
And ask one question that is not about tax at all: how would you want this presented if somebody reviewed it in three years. Accountants answer that well, because they have seen what an unexplainable schedule looks like from the other side, and the answer usually shapes the record keeping more usefully than any discussion of the rate itself does.
The same applies to the reasoning, not just the conclusion. A note saying three years is useful. A note saying three years because that is our actual refresh cycle, decided on this date, with the accountant’s name beside it, survives the person who wrote it leaving. That is the only real test of a policy.
Bring the four-field list with you. An accountant can answer these far more usefully when they can see what you actually own, and the exercise of producing the list frequently surfaces the real problem before the meeting starts.
Then write the answers down somewhere findable, with the date. Treatments change, people leave, and a decision nobody can locate gets made again from scratch in two years by somebody reading the same confident blog posts.
Put it with the asset register rather than in a finance folder. That is where whoever needs it next will look, and a policy stored where only its author thinks to look has the same practical effect as no policy at all.
HROpsLab publishes independent comparison work across HR and IT tooling. We sell nothing, we take no vendor money, and we publish no paid placements. Where a vendor publishes no price, we say so rather than estimating, and where a product is our own we state that too.
Frequently asked questions
What useful life does the BIR accept for a laptop?
No binding figure could be confirmed from a primary Bureau of Internal Revenue source, which is worth knowing given how many articles state one with confidence. Depreciation sits under Section 34(F) of the National Internal Revenue Code, and accounting practitioners report three to five years as the range commonly accepted for IT equipment. A frequently quoted 20 per cent annual rate is attributed to a regulation dating from 1940 that we could not verify as still operative, so treat the number as professional judgement rather than a published rule and confirm it locally.
Is straight line depreciation required in the Philippines?
Sources that do this professionally disagree, which is the honest answer. Some state that straight line is required under Section 34(F) as amended by the CREATE Act. Others say alternative methods are permitted where prescribed and that no specific rules have been issued on method or useful life at all. Both positions are published by firms practising in this area, so this is not a question to settle from an article. Ask a Philippine accountant what applies to your circumstances before adopting a treatment.
Can we change the useful life we use later?
Reportedly not freely, since changing a useful life or a depreciation method once adopted is described as requiring approval from the Bureau of Internal Revenue rather than being an annual choice. That matters more than it first appears, because it means the initial decision carries more weight than people assume when they pick a number quickly to get a schedule started. Choosing deliberately and documenting the basis at the outset is considerably cheaper than revisiting it.
Why does our refresh cycle not match the depreciation schedule?
Because they are set by different people solving different problems and nobody checks that they agree. Finance picks a useful life for the accounts. IT picks a replacement cycle based on workload, support cost and security support. When the book life is shorter you end up running fully depreciated machines that appear on no report, so the fleet ages invisibly. When it is longer, every refresh produces a disposal loss. Neither is a crisis and one conversation prevents both.
What does finance need from IT for the schedule?
Four things per machine, and IT usually does not know they are wanted: the purchase date, which is when depreciation starts and is too often approximated from an expense claim; the cost, including anything capitalised alongside the device; the serial number, which is the only identifier that survives a rebuild or a change of holder; and the disposal date with any proceeds. The fourth is where schedules decay, because machines leave the business and nobody tells finance, so the register slowly fills with equipment that no longer exists.
Should we keep a laptop that is fully depreciated?
Possibly, but do not let the accounting trigger make the decision, because a machine reaching the end of its book life has not stopped being useful and a machine carrying no book value has not stopped costing you anything. Replacing on the book schedule discards working equipment. Keeping indefinitely because it looks free ignores a rising failure rate and an operating system that will eventually fall out of support. Track age separately from book value and review the age distribution quarterly.
How should we handle a laptop we cannot recover?
Close the entry rather than leaving it open, which means a write-off with a stated approval threshold by device value so somebody actually has the authority to do it. Without that path, unrecoverable machines sit on the schedule indefinitely and the register gradually becomes fiction, which is worse than a smaller accurate one because people keep half trusting it. The accounting treatment of the write-off itself is a question for your accountant, since it depends on circumstances this page cannot know.
Does leasing change the laptop depreciation philippines picture?
It changes how the cost appears and when, since purchasing capitalises the asset and spreads it across the useful life while leasing or a subscription generally converts it to an operating expense. Whether that helps depends on your cash position and how you are measured, which makes it a finance question rather than an IT one. From the equipment side, the operational arguments usually outweigh the tax ones, and the comparison is often run badly because the purchase side omits spares, replacement logistics, recovery and time.
