Some exam questions are hard because the syllabus content is genuinely advanced. This one wasn’t. The real November 2024 Principles of Taxation Chief Examiner’s Report singles out a two-year corporate tax computation — the kind of question every candidate has practiced many times over — as “the worst attempted question in all.” This post walks through exactly what went wrong, using the real company, the real numbers, and the real marking scheme.
What the Report Actually Said
Question Four(a) of the real November 2024 paper gave candidates a short extract of Amasa Architecture and Building LTD’s operations for 2022 and 2023 — profit before tax, depreciation, two donations, capital allowances agreed with the Ghana Revenue Authority, and withholding tax paid — and asked them to compute chargeable income and tax payable for both years, worth 10 of the question’s 20 marks. The examiner’s verdict is direct: “The worst attempted question in all… Surprisingly, candidates did not grab the required marks. Almost all candidates failed to indicate the basis periods for the two years. Many showed a poor understanding of contribution to worthwhile causes (Donations which are regarded as allowed and not allowed). Some candidates were also subtracting withholding tax credits to arrive at chargeable income instead of deducting from gross tax payable.”
The Governing Principle
Three separate rules, each correctly known individually by most candidates, went wrong at three different points in the same computation. First, every year-of-assessment computation must open with a stated basis period — for Amasa, a company with a straightforward calendar year-end, that is simply “1 January to 31 December 2022” and the equivalent for 2023 — and the mark is available whether or not the basis period looks unusual. Second, donations must be tested individually against one question: is this an approved contribution to an approved cause, or not? The real scenario gave two donations that look identical on the page — both described only as “donation” — but the GH¢350,000 (2022) and GH¢210,000 (2023) to Manhyia Children’s Home was approved by the Social Welfare Department and treated as allowable, while the GH¢105,000 (2022) and GH¢150,000 (2023) towards the Adae Kese Festival had no such approval and was added back to profit in full, both years, in the real marking scheme’s computation. Third, withholding tax paid during the year is a tax credit, and a tax credit only ever reduces the final tax charged — it must never be subtracted anywhere in the build-up from profit before tax to chargeable income.
Where the Marks Are Lost
Each of the three errors damages the computation differently. Missing the basis period costs a small, specific, free mark that required no calculation at all — pure carelessness under time pressure. Misclassifying a donation cascades: get the add-back wrong for either year, and assessable income is wrong, which means the capital allowance of GH¢1,500,000 (2022) and GH¢1,700,000 (2023) is deducted from the wrong base, which means chargeable income and the resulting tax charged are both wrong. Misplacing the withholding tax credit corrupts the computation from the opposite direction: even a candidate who got every earlier line correct produces the wrong final “tax payable” figure if the GH¢10,000 (2022) or the combined GH¢35,000 (2023, made up of a GH¢10,000 overpayment carried from 2022 plus GH¢25,000 paid during 2023) is subtracted from chargeable income instead of from tax charged. Three ordinary, well-known rules, each contaminating a different part of a single 10-mark computation — that is exactly why the examiner called this the worst attempted question on the paper despite calling it, in the same breath, “a topical question.”
Exam Tip: Before touching a single number in a chargeable income computation: (1) write the basis period, (2) test every donation against “approved cause, approved authority?”, and (3) hold withholding tax paid back entirely until the very last line, where it reduces tax charged — never chargeable income.
Conclusion
This was not a knowledge failure — it was a sequencing failure, three times over, inside one question. Fix the sequence, not the knowledge, and a question the examiner named the worst attempted on the whole paper becomes one of the most reliable ten marks available on any future sitting.
Ready to Go Deeper?
This exact sequencing fix — and every other high-frequency computation trap examiners repeat sitting after sitting — is covered live inside our classes. Join a live class at Profs Study Mate.
For free weekly study tips and exam breakdowns like this one, join the Profs Tax Academy newsletter — and if you’d rather study from a physical textbook, our Taxation materials are available in hardcopy and digital formats, plus on encrypted pen drive for offline study.
