Tooling, and where the design work was done

The additions you invoice to yourself

Course 8.5 covers the ordinary buyer-supplied additions. Those are labels, hangtags and trims free-issued to the factory, and it shows how they get onto a value that shows none of them. Take that as read.

This lesson is about the two that are harder. Neither of them arrives as a purchase from the factory, and neither of them appears in a merchandiser's system at all. Both sit in the importer's own general ledger, filed under headings that have nothing to do with customs, and both were correct when they were filed there.

The Valuation Agreement requires the value of goods and services supplied by the buyer free of charge or at reduced cost, for use in producing the imported goods, to be added to the price. Its list has four limbs. The two that catch apparel importers are:

• tools, dies, moulds and similar items used in producing the goods; and
• engineering, development, artwork, design work, and plans and sketches necessary to produce the goods.

Read the second limb again when you get to it. It carries a condition the first one does not.

The moulds

OJ-217 carries a custom cord-lock and a custom zip puller in Peveril's own shape. Peveril commissioned the moulds from a toolmaker in Taiwan, paid USD 14,600 for the set, and had them shipped to Suvira, who uses them at no charge. Suvira never invoices Peveril for them, so nothing about them ever appears on a commercial invoice.

They are an assist. The USD 14,600 has to reach the customs value of the jackets those moulds produce.

The interesting question is not whether. It is how it is spread. The Agreement's own material accepts more than one answer, provided the method is documented and consistent with generally accepted accounting principles. Three are ordinary:

• All of it on the first entry. The whole USD 14,600 is declared on the first shipment, which is 18,400 jackets, so USD 0.7935 a jacket.
• Across the units produced to date, entry by entry, as the programme runs.
• Across the whole anticipated production. Peveril's tooling business case assumed 100,000 units over three seasons, so USD 0.146 a jacket, declared on every entry until the tool is exhausted.

Now the thing that makes this teachable. At the same duty rate, all three methods produce exactly the same total duty: 11.4% of USD 14,600, which is USD 1,664.40. Whatever you choose, the revenue is identical. What changes is only when you pay it. That is a cash question, not a compliance one.

Season one is 27,000 jackets: 18,400 knitted, 6,200 woven and the 2,400 that flew.

Where the third method goes wrong

Peveril chose the third, which is the normal choice and the cheapest in year one. Then the programme was cut. Peveril's autumn range was rebuilt after a poor season, and OJ-217 was dropped after two years at 51,000 units against the 100,000 the tool was costed over.

The moulds cost USD 14,600 and they produced 51,000 jackets. All of that cost relates to jackets that were actually imported, because there are no others. But Peveril declared USD 0.146 on each of them. That is USD 7,446.00 of assist and USD 848.84 of duty, against the USD 1,664.40 that the whole assist attracts.

USD 815.56 short, found, if it is found, in a post-clearance audit two years after the tool was scrapped.

The design split

The second limb is where the good story is.

OJ-217's development ran in two places, both correctly.

Peveril's own studio in Kessland did the concept, the colour work, the artwork and the range presentation, at USD 21,800 of internal cost. And because the fit block had been a problem for three seasons, Peveril sent the pattern engineering, the grading and the fit-block correction to Carvalho Pattern Studio in Porto, who are better at it, for USD 9,300. Carvalho's patterns were supplied to Suvira free of charge.

Now read the limb again. Engineering, development, artwork, design work, and plans and sketches are added only where they are undertaken elsewhere than in the country of importation.

So:

• The Kessland studio's USD 21,800 is not added. It was undertaken in the country of importation, and the Agreement carves it out.
• Carvalho's USD 9,300 is added. Same kind of work, done in Porto.

The distinction is geographic and it is deliberate. An importing country does not tax its own domestic design industry through the customs value of the goods that industry designs.

Peveril's finance system holds one line: OJ-217 development, USD 31,100. One cost centre, one project code, one number. That is exactly right for management accounting, and unusable for customs.

Two opposite errors, both reached honestly from the same true number, with a swing of USD 3,545.40 between them.

The mistake nobody made

Four decisions, all defensible, in four different departments:

1. Design used a Portuguese pattern house because it produced a better fit block. Correct.
2. Finance booked all development for one style to one project code. Correct, because that is what a project code is for.
3. The broker asked the right question: what did the buyer supply free of charge to the factory? Correct.
4. Somebody in merchandising answered it: the trims and the labels. Also correct, because in that person's world a pattern is not something you supply. It is something that exists.

Nobody was asked where the pattern work was done, because nobody knew the answer mattered. The question that unlocks it is not "what did we supply". It is "what did we pay for that ended up in the factory's hands, and in which country was it made?"

The assist register

One page per purchase order, kept by whoever owns the PO. Six columns:

1. What was supplied to the factory free or below cost.
2. What it cost the buyer, and the currency.
3. In which country it was produced or performed.
4. Whether it is consumed per garment or is a tool with a life.
5. If it is a tool, the apportionment method and the assumed production.
6. The date the assumption was last reviewed.

Column three is the one nobody thinks to include, and the one that decides the design limb. Column six is the one that would have caught the moulds.

Act as a customs valuation specialist reviewing an importer's own records for buyer-supplied additions to the customs value. Build me an assist register for one purchase order. The order: buyer [WHO AND IN WHICH COUNTRY], factory [WHO AND WHERE], style [CODE], quantity on this order [UNITS], total quantity expected over the life of the style [UNITS], price term [TERM]. Things I supply the factory free or below cost, one a line: [WHAT IT IS, WHAT IT COST ME, THE CURRENCY, WHERE IT WAS MADE OR PERFORMED, AND WHETHER IT IS CONSUMED PER GARMENT OR IS A TOOL WITH A LIFE]. Development spend on this style, split by where the work was physically done: [IN MY OWN COUNTRY: WHAT AND HOW MUCH] and [ANYWHERE ELSE: WHAT, WHERE AND HOW MUCH]. Do the following. First, sort every line into added to the customs value, not added, or needs a decision, and give the reason in one sentence each. Second, for the development spend, apply the rule that engineering, development, artwork, design work and plans and sketches are added only when undertaken outside the country of importation, and tell me plainly which of my lines that carves out. Third, for each tool, show me three lawful apportionments: all on the first entry, across units produced to date, and across the whole anticipated production. Give the value declared per unit and the duty effect of each, and say what changes and what does not. Fourth, tell me what happens to the un-apportioned balance if the style is dropped before the anticipated volume is reached, and put a review date on it. Fifth, list the questions I should ask my own finance and design teams to find lines I have not told you about, phrased so a non-customs person can answer them. Do not invent amounts. Where I have left a cost blank, list it as unquantified and tell me who inside my business holds the number.