La Bourse / Volume IV / Nº IV.07 / Workbook — the executive

A woman reading in a rattan chair beside a tall white window, light falling across her and the floor. — Plate IV.07 · Workbook — the executiveThe Second Room.Everyone draws a supply chain as a line because a line fits on a page. The thing itself is a population, and it is lit from a room you have not entered.

WORKBOOK — THE CORPORATE EXECUTIVE

Chapter IV.07 · Supply Chains as Ecologies

Volume IV — Production and Regeneration

Ninety days, applied to a business unit with a P&L and a board that will ask what it costs. Every figure below is computed in lib/verify/IV_07.py and can be re-run against your own inputs.

THE PREMISE, STATED COMMERCIALLY

You already spend money on supply risk. You spend it on qualification, on audits, on dual sourcing, on expediting, on the safety stock Chapter III.10 prices, and on the people who manage all of it. What you almost certainly do not have is the arithmetic that says whether each of those spends is buying anything.

This workbook produces four numbers for your unit, and each of them changes a decision:

The visibility bill — what it costs to see tier two and tier three, and what fraction you actually get.
The exposure ranking — which cheap components gate the most revenue, in an order that usually contradicts your spend report.
The source count per family — computed from what a stoppage costs you, which defends both single sourcing and quad sourcing where each is right.
The enforcement depth n* — how many tiers your code of conduct actually reaches, and the two levers that extend it.

The fourth is the one your board has never been given, and it is the one that turns an ESG commitment from a sentence into an engineering specification.

The worked unit below. Revenue $180m, external spend $108m, 22 part families, currently three qualified sources each — 66 direct suppliers. Replace these with yours; the shape of the answer does not change.

PART ONE — DISCOVERY

Days 1–30: find what you already have

1.1 — Pull the three lists you already own. Your AP ledger by vendor, your approved-vendor list, and your quality system's audit register. They will disagree. The disagreement is the first finding and it is usually worth writing down on its own: vendors paid but not approved, vendors approved and never used, audits registered against sites that closed.

1.2 — Find the relationships that have already saved you. Ask your plant managers one question: who went out of their way for us in the last three years? You are looking for recruitability, not redundancy — the Aisin property. Sixty-two firms made a brake valve they had never made, in five days, because a network already existed. You have some of that and it is on no register. Name the relationships. They are an asset you are currently not counting and, in several cases, actively eroding by re-tendering.

1.3 — Find your waist. Somewhere upstream your population narrows: one heat-treatment process, one certification body, one alloy mill, one port, one software licence, one accreditation. Ask your engineers rather than your buyers; engineers know where the narrow places are and buyers know where the money is, and those are different maps.

1.4 — Count what a stoppage costs. Not the part. The line. Contribution margin per day, plus expediting, plus customer penalties, plus the share of customers who do not come back. Get finance to sign this number. Every decision in Part Two is a function of it, and an unsigned event cost is how these analyses get argued away in month four.

PART TWO — THE ARITHMETIC

Days 31–45: four numbers for the board

2.1 — The population and the visibility bill. With 22 families × 3 sources = 66 direct suppliers and b = 5:

  tier 1      66 firms        $79,200/yr to map
  tier 2     330              $396,000
  tier 3   1,650              $1,980,000
  tier 4   8,250              $9,900,000

Tiers one to three: $2,455,200 a year, 2.27 percent of external spend, and at a sixty percent disclosure rate you see thirty-six percent of tier three. Take this table to the risk committee before they ask for full transparency, not after.

2.2 — The exposure ranking. For every family: the value of output it gates, divided by what you pay for it.

        E  =  gated output value  /  node revenue

In the chapter's reference case a $3.50 microcontroller gates a $35,000 vehicle — 10,000× — while the entire chip content of the vehicle gates it at 70×. Sort your families by E and print the list beside your spend list. They will disagree, and the disagreement is the board slide. Your risk register is currently sorted by the wrong column.

2.3 — The source count. For each family, the question is not "how many suppliers is prudent" but "what would a stoppage have to cost for the next source to pay for itself." With p = 0.05, q = 0.015 and $18,000 a supplier a year in qualification, audit and administration:

  1 -> 2   pays if a stoppage costs more than      $23,083,088
  2 -> 3   pays if a stoppage costs more than     $461,661,769
  3 -> 4   pays if a stoppage costs more than   $9,233,235,373

If your signed event cost is $40 million, two sources is right and three is a donation. If it is $600 million — a carmaker, a grid operator, a vaccine line — three is right and four still is not. The asymmetry comes entirely from q, the common-mode floor, and q is the only term you can genuinely reduce: move one source out of the shared region, off the shared port, away from the shared sub-tier.

2.4 — Enforcement depth. With 66 direct suppliers, $6,000 an audit, b = 5, λ = 0.20 (three sources, so your share of each supplier is modest) and a value of full compliance of $4,000,000 a year:

  C₁ = 66 × 6,000 = $396,000
  V₁/C₁ = 10.1010          ln = 2.3126
  b/λ   = 25.0000          ln = 3.2189
  n*    = 1 + 0.7185  =  1.718 tiers

Your code of conduct reaches one and seven-tenths tiers. Now the lever: buying inspections jointly with seven other buyers takes c to $750, so C₁ = $49,500, V₁/C₁ = 80.808, and

  n*  =  1 + 4.3921 / 3.2189  =  2.364 tiers

Plus 0.646 tiers, and it is cash-positive on day one: cancelling duplicate audits on 66 suppliers saves 66 × $5,250 = $346,500 a year. A reach extension that pays you $346,500 to take it is not an ESG initiative. It is a procurement saving with a governance side effect, and it should be presented in that order.

PART THREE — DESIGN

Days 46–60: the instrument

3.1 — Join a group inspection scheme. The Accord, the Responsible Minerals Initiative, Sedex, amfori BSCI and their industry equivalents already exist. You are joining, not building. Budget one quarter and a legal review; the antitrust question is well-trodden because the schemes were designed with it in mind, and inspection of labour and safety conditions is not commercially sensitive information about price or capacity.

3.2 — Cancel the duplicates. Every supplier audited by the scheme comes off your private audit programme except the twenty relationships where you have real leverage and a genuinely higher standard. Keep that thin private layer deliberately: a group standard is a floor and will be lower than what your best suppliers already do.

3.3 — Move the enforcement to the waist. Where your population narrows, enforce there. The chapter's mineral case: 8,000 sources through 300 smelters, auditing at the waist costs $1.8m instead of $48m — a 96.25 percent saving — and shared across eight buyers is $225,000 against a tier-three compliance value of $270,000. It pays at exactly the tier the geometric arithmetic calls unreachable.

3.4 — Build the deep-tier financing programme. This is the mechanism that reaches tiers your contracts do not.

Structure. Your bank pays approved supplier invoices early against your credit rating. The supplier is paid at day five instead of day sixty-five at a discount set by your risk, not theirs.
Two rates. A base rate for approved suppliers; a lower rate for suppliers holding a current verified standard. The differential is the enforcement mechanism and it costs you nothing — it is the bank's pricing.
Nomination. Tier-one suppliers nominate their own critical suppliers into the same facility, which is how the programme reaches tier two and three.
Balance sheet. Trade payables stay trade payables only while commercial terms are unchanged and participation is voluntary. Extend terms as a condition of access and the auditors reclassify the balance as borrowings. Disclose programme size and payment terms — the IAS 7 and IFRS 7 supplier-finance disclosure requirements are now explicit, and a reclassification discovered at year end is how these programmes die.
Counterparty. Your existing cash-management bank first. A single fintech platform holding the rails for your deep tiers is a new node with a very high exposure ratio; treat it as one.

3.5 — The number that decides it.

  supplier's own working-capital rate     12.00 %
  rate against your credit                 5.00 %
  spread                                    700 bp
  payables financed                       60 days

  benefit per dollar of invoice = 0.07 × 60/365 = 1.1507 %

  breakeven annual trade = verification cost / 1.1507 %
       solo verification  $6,000  ->  $521,429
       shared             $750    ->   $65,179

Above the breakeven a supplier pays for its own verification out of the financing spread, and the standard needs no enforcement at all. In the chapter's reference case tier-two nodes average $150,000 of trade — below the solo line and well above the shared one, which is precisely why joining the group scheme is the move that makes the financing programme reach.

PART FOUR — DESTINY AND DELIGHT

Days 61–90: make it hold

4.1 — Put four numbers in the monthly pack. Enforcement depth n*. Tier-two disclosure rate. Number of families whose source count is justified in writing. Percentage of spend covered by a dated map, with the uncovered remainder stated. Anything reviewed monthly persists; anything reviewed by exception does not.

4.2 — Give the map a second owner. One named owner and one named deputy, both in the operations review. A map maintained by one enthusiast decays at the rate that person changes roles.

4.3 — Date every node. A node not re-verified within twelve months reads as unknown, never as unchanged. Build this into the report rather than into a policy, because a report enforces itself.

4.4 — Present it to the board in this order. The saving first ($346,500 of cancelled duplicate audits). The reach second (1.718 to 2.364 tiers). The resilience third (the source-count table with the signed event cost). Never in the reverse order: a governance argument that opens with a cost is heard as a cost, and this one is not a cost.

On delight. The satisfying moment is not the board approval. It is the first incident after the map exists, when the question which families are exposed to this is answered in an afternoon instead of a fortnight and somebody says so out loud. The second is when a competitor's procurement director agrees to share an inspection and you both realise you have been paying twice for the same visit to the same factory for years. That conversation takes ten minutes and the saving is permanent.

THE FAILURE MODES, NAMED

So you can see them coming

The map goes stale silently. Suppliers change their own suppliers without telling you. Dating every node is the whole defence.

The exposure ranking is re-sorted by spend. It happens the first time the list is handed to someone who was not in the room. Write the ratio into the column header so the ordering explains itself.

The group standard becomes the ceiling. Keep the thin private layer for the twenty relationships where you have leverage. This is the one duplication worth paying for.

The arithmetic is used to justify stopping. n = 1.718 is a fact about a configuration, not a law, and it is trivially misread as upstream conduct is not our problem*. The same equation says shared inspection plus partnership depth reaches 2.98 and the waist reaches further. Quote both halves or you are using arithmetic as an alibi.

The financing programme gets reclassified. Named above because it is the one that costs real money if missed. Involve the auditors in month two, not month eleven.

THE NINETY DAYS ON ONE PAGE

Day	Action	Artifact
1–15	Reconcile AP, AVL and audit register; sign the event cost	The signed event cost
16–30	Exposure ratio for every family; find the waist	Ranked list, beside the spend list
31–45	Map the top decile to exhaustion, every node dated	The map, three pages
46–60	Join a group inspection scheme; cancel duplicates	The subscription, and $346,500
61–75	Two-rate facility agreed with the cash-management bank	Term sheet, auditor sign-off
76–90	Tier one onboarded, tier-two nomination open	First tier-two supplier funded

BOARD PAPER TEMPLATE

One page. Six lines.

What we found. Our code of conduct is enforceable to a depth of 1.718 tiers. We have been describing it as applying throughout the supply chain.
What it costs to see further. Tiers one to three cost $2,455,200 a year and return thirty-six percent of tier three. We do not recommend it across all families.
What we recommend instead. Map the top decile by exposure ratio. Join the group inspection scheme. Move enforcement to the waist.
What it saves. $346,500 a year in cancelled duplicate audits, from day one.
What it buys. Enforcement depth from 1.718 to 2.364 tiers, and a financing route that reaches tier two without a contract there.
What it does not cover. Ninety percent of families remain mapped to tier one only. That is a deliberate choice and here is the arithmetic behind it.

Line six is the one that gets the paper believed.

APPRECIATIVE QUESTIONS FOR YOUR LEADERSHIP TEAM

When has a supplier protected us during a shortage when they did not have to? What had we done, over the years, that made that happen?
Which of our smallest suppliers gates the most revenue — and who in this room already knew the answer without a system?
If we could answer "who makes this, and where" in a day rather than a month, what would we decide differently first?
Which competitor would we call about sharing an inspection, and what is the one sentence that gets a yes?
What would we want to be known, in ten years, for having done first in this industry — and what part of it could we start this quarter?