An invented supplier ethics case: capacity arithmetic, an undisclosed site, and a fast exit

The company below does not exist and none of this happened. It is an invented mid-sized manufacturer of consumer goods, written so a supply chain problem can be followed from the first odd number to the decision that closed it.

It is included because the company did most things right and got the last decision wrong, which is the more common and less discussed failure.

What to take away

Week one: a number that did not reconcile

An analyst in the sourcing team was doing an ordinary capacity review before agreeing the next year's volumes. She compared the declared workforce and shift pattern at a supplier's site against what the company had received over the previous twelve months.

The site could not have made it. Not marginally: the received volume implied roughly half again as much labor as the declared headcount could provide, even assuming continuous operation.

She checked her own working first, which is the right instinct and cost two days. The gap held. She raised it with the sourcing manager in writing, with the arithmetic attached.

Weeks two and three: the answers that did not fit

The supplier's first explanation was a productivity improvement. The second, when the numbers were put to them again, was seasonal agency labor that had not been included in the declared headcount.

Both explanations were plausible in isolation. Neither survived a follow-up question. Agency labor of the scale implied would have appeared in the payroll reconciliation the company already received, and it did not. The productivity claim would have shown in the equipment list, and the equipment had not changed.

At this point the company did something that deserves credit: it stopped asking the supplier and went to look. It exercised a short-notice access right that had been negotiated at contract stage two years earlier and never used.

Week four: the second site

The visit found a second production site, twenty minutes away, operated by a subcontractor, doing roughly forty per cent of the company's volume. It had never been disclosed, had never been audited, and was not covered by any of the certificates in the file.

Conditions there were worse than at the audited site. The specifics do not matter for the lesson and, being invented, would only distract. What matters is the structure: three announced audits over two years had reported no findings, and all three had been conducted at the site that was arranged to be audited.

The company's own published statement described its supply chain as fully audited to tier one, a description that had been true of the sites it knew about. That is the failure mode of every such claim.

Where it applies, the duty to describe supply chain due diligence honestly makes publishing an annual modern slavery statement a discipline, not a formality. What any company must publish is a question of local law and counsel.

Week five: the decision that went wrong

The finding reached a director on a Tuesday. The relationship was terminated the following Thursday, nine days after the second site was confirmed.

The reasoning was not unreasonable on its face. The undisclosed subcontracting was a clear breach. The conditions were unacceptable. The company had a published position and an approaching reporting cycle. Legal advice was that terminating was available under the contract.

What nobody did was ask who would bear the cost of the exit. The subcontractor's site employed several hundred people and had one significant customer at one remove. When the orders stopped, the site stopped, and a substantial number of workers were owed wages for work already done.

The contracting supplier disclaimed responsibility on the basis that the subcontractor had never been authorized. That was true, and it was also the company's own finding being used against the people it was supposed to protect.

The company had ended its exposure in nine days and had made the situation of the workers worse. Both of those things were true at once, and the second was not discussed in the meeting that decided the first.

What the company should have compared

Exit was one of three options and the only one considered.

Stay with conditions. Recognize the subcontractor, bring it inside the program, set specific requirements with dates and verification, and make the next order conditional on them. Costly, slow, and the only option under which the workforce keeps its income while conditions change.

Exit responsibly. Terminate on a timetable rather than immediately, with the wage liability secured as a condition of the final payments, and with notice long enough for the site to seek other work.

Exit immediately. What happened. It is the correct choice when there is a risk of serious harm now, and this was not that situation. The risk being managed was the company's.

The comparison takes an afternoon. It was not made because the decision was framed as compliance rather than as a choice between harms, and because the person deciding was measured on the company's exposure and on nothing that happened afterwards.

What changed

A written decision rule about exit. Agreed and signed off before the next finding: immediate danger moves on its own clock; everything else requires a written comparison of staying with conditions, exiting on a timetable, and exiting at once, with the cost to the workforce estimated in each. A rule drawn while a shipment waits is not a rule, which is the general point made in ethical dilemmas.

Survival of payment obligations. Contracts were amended so that termination does not extinguish amounts owed for work performed anywhere in the chain, with flow-down that has to be evidenced rather than asserted.

Capacity arithmetic as a standing check. Run annually against every supplier above a volume threshold, using data the company already had. It is now the check that has found the most, and the honest reporting of what such checks establish is the subject of corporate responsibility.

A worker-facing channel. Built in the following year, reaching a person at the company rather than at the supplier. It produced reports immediately, which the program was warned in advance to expect rather than to read as deterioration. What legal protection a worker who raises something has is set by the jurisdiction rather than by the buyer's policy, and the UK's whistleblowing rules are one example of such a regime.

What the case shows

Audits found nothing for two years and arithmetic found it in a week, because arithmetic cannot be prepared for. That is the finding most worth taking.

The second is harder. A company that finds a problem and exits fast has protected itself and has not necessarily helped anyone. The difference shows only if someone asks who pays for each option before the decision is made.

Who is placed to ask that, and why they usually are not, is set out in leadership ethics. The wider argument about what a buyer can require, fund and see is in supplier ethics.

Common questions

Was terminating the wrong decision?

Not necessarily. Terminating without comparing it against the alternatives, and without securing the wages owed, was the failure. The same outcome reached deliberately, on a timetable, would have been defensible.

Could the audits have caught this?

Short-notice visits might have. Announced audits at a site chosen by the supplier could not, and reporting their passes as assurance is what allowed two years to elapse.

Is the capacity check realistic for a smaller buyer?

Yes, and it is the cheapest thing on this site's supplier tools list. It needs declared headcount, shift pattern and your own receipts, all of which you already hold.