Counting desks while the useful life goes unexamined

Anyone who has watched a public sector accrual conversion up close has seen the same scene. A team with clipboards, or a tablet and a barcode printer, working through an office floor recording desks, chairs and monitors. Each one gets an identifier, a classification, a cost and a useful life. Somewhere in the same organisation sits a property portfolio worth 400 times as much, carried on a single straight line.

The usual objection is that the first activity is a waste of money. It is, but that is the weaker complaint, because a waste of money can be defended as prudence and usually is. The stronger objection is that the standard did not ask for it.

What the standard actually asks for

Read IPSAS 45 for its modal verbs and the picture inverts.

On individually insignificant items the standard is permissive. It declines to prescribe a unit of measure for recognition at all, says judgement is required, and offers aggregation as an option, naming library books, computer peripherals and small items of equipment as its examples. Nothing there compels an entity to give a monitor its own line and its own depreciation schedule.

On significant parts it is not permissive.

Three requirements, each carrying the word shall, and each concerned with the assets where the money is.

Four treatments of a fixed asset register set against what IPSAS 45 requires. Capitalising individually insignificant items is permitted rather than required, and paragraph 8 offers aggregation instead. Separate depreciation of significant parts, and annual review of useful life, residual value and depreciation method, are all requirements. Effort in practice runs the other way.

Treatment	What the standard says	Where the effort goes
Capitalise low value items individually	Not required. Para 8 permits aggregating them	Most of it
Depreciate significant parts separately	Shall, para 41	Rarely attempted
Review useful life and residual value	Shall, annually, para 48	Often a rollover
Review the depreciation method	Shall, annually, para 58	Almost never

The one row the standard leaves to judgement is the one that gets industrialised. The three carrying the word shall are the three that depend on knowing how long things last, which is a question the finance function cannot answer alone and rarely asks anyone else.

Put the two asset classes on the same page

The quickest way to see the imbalance is to write both classes out the way a financial statement writes them, in millions, and look at what the tagging exercise is protecting.

That last figure is the one to hold. The entire annual depreciation of every desk, chair and monitor the entity owns moves the balance sheet by 0.036%. An error of 10% in it is 0.004%. No audit materiality threshold reaches down that far, and none is meant to.

Now look at the other side of the same statement. The 400m charge rests on one assumption, the 50-year life. Shorten it by a decade and the charge becomes 500m, a movement of 100m a year. That is 14 times the entire office equipment charge, and it turns on a single judgement that paragraph 48 requires somebody to revisit annually.

And componentisation, which is the same argument one level down

A building is not one thing with one life either. Structure, services and fit-out wear out at different rates, which is the entire reason paragraph 41 exists. Applying one life across all of them is not conservative. It is a different answer.

Set the 190m beside the 7.14m and the priority is not arguable. One is a recurring annual difference on the largest class in the accounts. The other is the whole depreciation of the furniture.

Why the effort goes to the wrong end

Not stupidity. The tasks are not equally answerable.

Tagging is visible, countable and terminates. There is a floor, a number of items on it, and a moment when the floor is done. It can be resourced with temporary staff, tracked as a percentage, reported to a committee and audited for completeness. Every property it has as a piece of work is attractive.

Deciding whether the structure of a building has a 50-year life or a 60-year one has none of those properties. It has no visible artefact, no completion percentage, and it cannot be done by the finance function alone, because the answer depends on construction, condition and intended service, which live in a technical function that in many entities does not yet exist.

Why the technical view is usually the one that has not been built

, read: An entity is asked for four asset registers and builds oneAn entity is asked for four asset registers and builds one

So an organisation faced with two obligations does the one it can complete. The reporting then shows a register at 98% completeness, which is true, and says nothing at all about whether the numbers in it are right.

The audit question this raises

If the register is complete and the useful lives are unexamined, an audit that tests completeness will pass an account whose largest estimate has never been challenged.

That is not a criticism of testing completeness, which is a real assertion and has to be covered. It is an observation about balance. The existence and completeness of a fixed asset register is a records question. Whether the depreciation charge on the largest asset class reflects the consumption of service potential is a judgement question, it is where the number actually comes from, and paragraph 58 requires somebody to revisit it every year.

What to examine

• Ask when the useful lives were last reviewed, and what changed. If the answer is that they were set at conversion and carried forward, the annual review required by paragraph 48 is not happening, whatever the policy says.
• Ask whether any building is componentised. Then ask whether the components were chosen for depreciation or for maintenance, because the two breakdowns rarely coincide and only one of them is in the ledger.
• Ask what the capitalisation threshold is, when it was set, and against what reasoning. A threshold inherited without review is not a judgement, and paragraph 8 assumes a judgement was made.
• Compare the cost of the last inventory exercise with the value of the assets it added to the register. Where the first exceeds the second, the exercise has consumed more than it recorded.

---

Sources. IPSASB, IPSAS 45, Property, Plant, and Equipment, effective for periods beginning on or after 1 January 2025, replacing IPSAS 17. Paragraph 8 for the unit of measure, the requirement for judgement and the option to aggregate individually insignificant items, with library books, computer peripherals and small items of equipment as the standard's own examples. Paragraph 41 for separate depreciation of parts significant in relation to the total cost or value. Paragraph 48 for annual review of residual value and useful life. Paragraph 58 for annual review of the depreciation method. Basis for Conclusions BC28 and BC29 for the decision to leave capitalisation thresholds to management applying users' information needs, materiality and cost against benefit. Published free by IFAC as part of the IPSAS handbook.

Ministry of Finance, Kingdom of Saudi Arabia, The Frequently Asked Questions Guide for the Asset Inventory, Stocktaking and Valuation Initiative in Government Entities, first edition, June 2024, for the treatment of below-threshold items as inventory expensed on issue and tracked in a control register, and for the instruction to weigh the estimated cost of a valuation exercise before undertaking it. Published in Arabic; the ministry maintains a public FAQ page for the programme.

The componentisation arithmetic is this publication's own illustration. Its assumptions are stated in full where it appears so that a reader can substitute their own and see what changes.