HMRC business economics exercises

Q: What does a simple model look like?

Illustration only. A shop's accounts show purchases of £80,000 and sales of £92,000. The trader agrees from invoices that the average mark up is 25%, which converts to a gross profit rate of 20% using the EM3503 formula (25 divided by 125, times 100). Theoretical sales: £80,000 plus 25% is £100,000, a gap of £8,000. The trader shows from supplier credit notes and disposal records that £4,000 of stock at cost went out of date. Sales on the remaining £76,000: £95,000. Goods costing £1,000 were taken for the family and not adjusted in the accounts. Sales on the remaining £75,000: £93,750. Once wastage and own goods are evidenced, the gap falls from £8,000 to £1,750. The trader can then argue that the remaining difference is within the margin of error of a sampled mark up.

Q: How should you respond to a business economics exercise?

Ask for the workings and assumptions. EM3502 says HMRC should state every assumption, such as wastage or sales per container, and draw attention to anything estimated. Check the sample period. If it may not be representative, EM3502 says the officer should consult the taxpayer and extend it or choose another period as a cross check. Evidence your explanations. Price cuts, wastage, theft, competition and inexperience are all explored in EM3508, but EM3505 expects corroboration such as price lists and examples. Offer a fuller analysis. In the EM3508 example the trader was invited to redo the calculation on a full year of invoices using the agreed mark ups. Use the dispute routes. EM3502 points to alternative dispute resolution where there is an impasse (see HMRC ADR). Otherwise HMRC can amend the return under section 28A TMA 1970 by closure notice, and the taxpayer may appeal (EM3508). For VAT, a model of this kind is often the basis of an assessment to best judgement under section 73 VATA 1994, as in Karoulla; see VAT best judgement assessments. If HMRC opens a check into your return, tax investigation insurance pays your accountant's fees. Solar Protect cover pays up to £100,000 with no excess; it never pays tax, interest or penalties. Read how our cover works or get a quote.

A business economics exercise is HMRC’s reconstruction of what a business should have taken, built from its purchases, prices, mark ups and wastage, and sometimes from test purchases, then compared with the takings declared. HMRC’s own Enquiry Manual says a model on its own is not conclusive: it carries real weight only once the records have been shown to be unreliable, and its figures should come from the facts of that business rather than an industry average.

Tax investigation insurance from £108.75 a year (£9.06 a month).Pays up to £100,000 of your accountant's fees if HMRC opens an enquiry. No excess. Buy online in a few minutes.

See my price in 2 questions How the cover works Accountant? Offer cover to your clients or call 01795 841 310

What is a business economics exercise?

HMRC’s Enquiry Manual calls it a business model (EM3502). The officer applies formulae to the Standard Accounts Information in the return, to the accounts or to the underlying records, arrives at an estimated turnover, and compares it with the figure declared. The model used most often in HMRC Direct Taxes is based on gross profit rate, but EM3515 says whatever goes into a business can be used as the base, provided there are reasonably accurate figures for what is put in and for the selling prices of the goods or services. A price out per hour model, for example, needs hours worked, chargeable hours and charge out rates (EM3505).

The courts have accepted business models as a legitimate way of calculating profits, and HMRC cites two cases. In Brittain v Gibb (59 TC 374, EM3710) the Commissioners estimated a painter and decorator’s profits by multiplying his likely hourly charge by the hours in his likely working year. In Coy v Kime (59 TC 447, EM3711) a London cab driver’s notebook, which consistently showed daily takings of £24 or £25, was found not to be an accurate record, and the Inspector’s fuel to takings model was held to be admissible evidence.

Why do mark ups and gross profit rates attract attention?

EM3082 describes gross profit rate (GPR) as the most commonly used business ratio in HMRC: gross profit divided by turnover, times 100. Officers compare it with other businesses, nationally or locally, and with earlier years. A lower than average rate, a widely fluctuating rate in a trade where terms do not vary much, or a substantial rise or fall in one year may each be treated as a risk indicator. Mark up is the margin on cost rather than on sales, and EM3504 warns that the two are easily confused; see gross margin vs markup for the conversion.

The ratio is a blunt tool in larger businesses. EM3082 gives the example that on turnover of £200,000, omitting £2,000 of sales changes the gross profit rate by only 1% of sales, so in bigger cases HMRC usually has to demonstrate specific errors in the records.

Worried about an HMRC check? Tax investigation insurance pays your accountant’s fees if HMRC opens an enquiry or compliance check. Solar Protect cover pays up to £100,000 with no excess. It never pays tax, interest or penalties.

From £108.75 a year (£9.06 a month) for individuals, £117.57 for sole traders and £207.50 for companies and partnerships, including IPT and the £20 admin fee.

See my price in 2 questions or call 01795 841 310.

★★★★★ Rated 4.9 out of 5 on Trustpilot from over 160 reviews

How does HMRC build a mark up model?

EM3505 says it will nearly always be necessary to meet the trader, preferably at the business premises, to understand how the business runs. For a gross profit model the officer needs cost, volume, the mix of purchases and pricing policy. In practice the officer will:

examine a sample of purchase and sales invoices to establish mark ups and product mix, trying to make the sample representative and allowing for seasonal trade (EM3505);
agree a mark up for each category of goods, relying on invoices, current costs and selling prices, and weight them by the purchase mix (EM3508);
reduce the theoretical maximum for wastage and pilfering, which the trader is expected to be able to quantify at least approximately (EM3505 and EM3509);
adjust for goods taken for own use, depending on how they were treated in the accounts (EM3510);
check whether the result fits what is known of the trader’s private lifestyle (EM3520).

EM3502 lists the errors that commonly undermine a model: failing to adjust correctly for VAT, stock and own goods, ignoring price changes during the year, and not taking proper account of discounts and loss leaders that change the product mix.

Where do test purchases fit in?

In cash businesses HMRC may arrange test purchases and observations, then check whether those sales appear in the till records. The Upper Tribunal decision in Karoulla t/a Brockley’s Rock v HMRC [2018] UKUT 0255 (TCC), a VAT appeal about a shop, shows how that evidence is used and where it can fail. HMRC relied on sales recorded while the till was in training mode, card sales missing from the end of day Z readings, and purchases and observations by test purchasers that were not reflected in the Z readings. The suppression rate of 16.68% used in the assessment was the average of the training mode figure (18.39%) and the test purchase figure (17.74%).

The First tier Tribunal gave the test purchase evidence less weight because HMRC did not produce the till rolls on which it was based. On appeal, the Upper Tribunal found that the officer’s conclusion that card sales were left off the till after 8pm every day rested on the till rolls for only three days, although HMRC had taken rolls covering around a dozen days. The other rolls showed the suppression was inconsistent. The Upper Tribunal admitted that new evidence, set the decision aside and sent the case back for a fresh hearing. The point for any business is that a test period must be representative, which EM3502 also requires. Where till manipulation is suspected, see electronic sales suppression.

Can HMRC rely on industry averages?

Only in limited ways. EM3502 tells officers to use facts from the actual business, preferably from contemporary documents such as price lists and purchase invoices, and not to use an average mark up, national or local, unless there is no evidence at all of the particular mark up. Any average used must be made clear in the computation. EM3082 says reconstructing profits from the mean rates achieved by apparently similar businesses should only be done where there is no cooperation from the trader.

Averages are weak evidence before a tribunal. EM3082 accepts that evidence about other businesses is not evidence about the taxpayer’s business, that a summary of other traders’ rates is hearsay and probably a breach of confidentiality, and that an enquiry officer is unlikely to be accepted as an expert on a particular trade. Benchmarks can justify selecting a case and can inform a best judgement figure, but the tribunal needs evidence that this taxpayer’s profits are inadequate. EM3711 notes that Coy v Kime shows averages may have some place when actual figures are not available; there the Inspector supported his fuel model with a Family Expenditure Survey of likely living costs.

Does a model prove that takings were understated?

Not by itself. EM3502 says that in isolation a model will not provide a conclusive figure to overturn the return. If HMRC has discredited the underlying records, the model provides a plausible computation of the real profits; if there is no other reason to doubt the records, the model is merely an alternative and the tribunal might find the return more reliable. EM2701 makes the same point: showing that records are inadequate does not prove omissions, and the officer still needs evidence of their own to substitute a higher figure.

Once HMRC amends a return or raises an assessment, the onus of displacing it rests with the taxpayer. EM3520 reminds officers that this does not let them use whatever figures they like: the figures must be best judgements. In Johnson v Scott (52 TC 383, quoted at EM3705) Walton J said the inferences, and the figures themselves, must be fair. If an enquiry year is settled on a model, HMRC may use it to argue for earlier years under the presumption of continuity.

What does a simple model look like?

Illustration only. A shop’s accounts show purchases of £80,000 and sales of £92,000. The trader agrees from invoices that the average mark up is 25%, which converts to a gross profit rate of 20% using the EM3503 formula (25 divided by 125, times 100).

Theoretical sales: £80,000 plus 25% is £100,000, a gap of £8,000.
The trader shows from supplier credit notes and disposal records that £4,000 of stock at cost went out of date. Sales on the remaining £76,000: £95,000.
Goods costing £1,000 were taken for the family and not adjusted in the accounts. Sales on the remaining £75,000: £93,750.

Once wastage and own goods are evidenced, the gap falls from £8,000 to £1,750. The trader can then argue that the remaining difference is within the margin of error of a sampled mark up.

How should you respond to a business economics exercise?

Ask for the workings and assumptions. EM3502 says HMRC should state every assumption, such as wastage or sales per container, and draw attention to anything estimated.
Check the sample period. If it may not be representative, EM3502 says the officer should consult the taxpayer and extend it or choose another period as a cross check.
Evidence your explanations. Price cuts, wastage, theft, competition and inexperience are all explored in EM3508, but EM3505 expects corroboration such as price lists and examples.
Offer a fuller analysis. In the EM3508 example the trader was invited to redo the calculation on a full year of invoices using the agreed mark ups.
Use the dispute routes. EM3502 points to alternative dispute resolution where there is an impasse (see HMRC ADR). Otherwise HMRC can amend the return under section 28A TMA 1970 by closure notice, and the taxpayer may appeal (EM3508).

For VAT, a model of this kind is often the basis of an assessment to best judgement under section 73 VATA 1994, as in Karoulla; see VAT best judgement assessments.

If HMRC opens a check into your return, tax investigation insurance pays your accountant’s fees. Solar Protect cover pays up to £100,000 with no excess; it never pays tax, interest or penalties. Read how our cover works or get a quote.

Related Accountapedia entries

Sources

Reviewed by Christine Anderson, Broker, Solar Protect. Last reviewed 19 October 2026. Part of Accountapedia, the Solar Protect guide to UK tax and accounting. This page is general information, not personalised advice.