Lessons · Lesson 6 of 6
Deliberate, priced, or unnoticed
Separate the risk you are paid to carry from the risk nobody chose, and put the whole book on one page with six numbers.
Lesson 6 of 6 · 16 min
The boundary of the insurable world
There is an old and useful difference between two things. The first is a risk you can measure: the possible outcomes are known well enough that somebody will quote you a price to take it away. The second is an uncertainty you cannot measure: no price exists, because nobody can say how likely each outcome is. The first becomes a cost line in your accounts. The second stays with you, and it is the reason a factory earns a profit at all rather than a wage.
Zerrand's autumn book sits on both sides of that line. Name the three courses that own the first side, so that the second side is not mistaken for them.
- The container is lost overboard. Measurable, insurable, priced — course 12.8.
- The buyer takes the goods and does not pay. Measurable, insurable, priced — course 13.5.
- The rate moves between the quotation and the money. Measurable, hedgeable, priced — course 13.6.
- The buyer decides it does not want as many jackets. Nobody will quote you a price for that.
Say the line out loud, because it is the sentence this whole course turns on. An insurer pays when a buyer cannot pay. Nobody pays when a buyer does not buy. Every instrument a factory owns sits on one side of that line. The risk that took 58.4% of Zerrand's autumn margin sits on the other.
Three kinds of risk on one book
Everything on the uninsurable side falls into three groups, and only the third is a problem.
Carried deliberately, and paid for. Ternhill's workwear contract requires Zerrand to hold six weeks of finished stock ready for the buyer to call off. So the factory carries goods it has made and not sold. That is a real risk, and it is in the price: the contract carries a stated 1.9% for it, which is USD 22,610 across the order. Somebody spotted it, priced it, sold it and wrote it down. This is what a factory is for.
Priced, and refused. In February a buyer asked Zerrand to buy and hold 40,000 metres of an exclusive print with no deposit against it. The cloth commitment was USD 214,000. The order's costed margin was USD 61,300. So the order was worth 28.6% of the money it was asking the factory to put on the table. Zerrand asked for a 50.0% fabric deposit. The buyer said no, and Zerrand said no to the order. A refusal is a legitimate answer, and this one came from a division rather than a feeling.
Carried because nobody noticed. The 75.64% of the book in one group. The 26 line-weeks held with no expiry date. The single delivery window that put every affected order at the same stage on the same day. Nobody decided to carry any of it. Nobody was paid for any of it. And it cost more than the other two groups combined.
The remedy for the third group is not more caution. Caution is what produced the 35% buyer policy that passed. The remedy is a page.
The order-book risk sheet
Six lines. Each line has a number, a limit taken from the factory's own record rather than from a textbook, and a person who owns it. Zerrand built this after September and it took forty minutes, because every figure on it already existed somewhere in the building.
| Line | Limit | The autumn book | |
|---|---|---|---|
| Largest buyer's share of costed margin | 30.0% | 29.66% | pass |
| Largest group's share of value | 45.0% | 75.64% | breach |
| Effective number of groups | 3.00 or more | 1.63 | breach |
| Committed cost in the building on the busiest Monday, against the season's costed margin | 2.00 times | 2.59 times | breach |
| Held capacity as a share of the window, at conversion value | 8.0% | 11.4% | breach |
| Margin lost in a single bad week, under the stress rule | 25.0% | 60.1% | breach |
Five breaches out of six, on a book that passed every acceptance rule the factory had, and was presented to the board as its best autumn ever.
Two things about that sheet are worth more than the limits on it.
The first line passes, narrowly. Ospreyfield at 29.66% of margin against a 30.0% limit is a real result, and a sheet that only ever says no gets ignored. It also shows the value of weighting by margin: on turnover Ospreyfield is 22.00% and would not have come close to the limit.
The fourth line is the one that will surprise a merchandiser. Committed cost in the building against the season's costed margin asks a simple question. How many times over is your whole year's profit standing on the floor as cloth and unfinished garments? On 8 September the answer was USD 1,842,000 against USD 711,218, which is 2.59 times. A factory running at that ratio does not need a bad season to be in difficulty. It needs a bad fortnight.
The factory that cannot spread its book
Everything in lesson 5 assumes a factory with a choice, and plenty of factories do not have one. A single line making one product for two customers cannot build a portfolio. Telling it to spread its book is advice it cannot act on.
Its tools are different, and they are all about terms and price rather than spread.
- A fabric deposit, so the cloth commitment is not the factory's alone.
- A shorter quotation validity, so the price is firm for less time.
- A staged fabric commitment, buying cloth in two releases against a confirmed programme rather than one release against a forecast.
- A cancellation clause written against the stage rather than a flat percentage of order value. A flat percentage is right at exactly one point in production and wrong everywhere else.
The arithmetic in lesson 3 says which of those to fight for. Of Zerrand's USD 415,568.53, the margin given up was USD 98,709.00 and the committed cost it could not recover was USD 316,859.53. A better cancellation fee argues about the first figure. A clause that decides who owns the cloth the moment it is cut argues about the second, and the second is more than three times the size. Course 14.2 owns how each of those is drafted and what a buyer will actually sign.
What you should not try to remove
One caution to close on, because a course about risk can leave a reader wanting to remove all of it.
The uninsurable part of an order book is not a defect. It is the reason a factory is paid more than a subcontractor doing the same sewing for a fee. Imagine a factory that took no demand risk at all: no capacity held open, no cloth bought before the final call-off, no price fixed before the season. Nobody would have any reason to pay it a margin, because it would be carrying nothing on the buyer's behalf.
The aim is not a book with no risk on it. The aim is a book where every risk is in one of the first two groups: carried deliberately and paid for, or priced and refused. The third group is the only one that has to disappear.
Check yourselfA factory's largest buyer is 24.0% of turnover, it insures its ledger, it hedges its currency and it has a cargo policy. Its finance director says the order book carries no uncovered risk. What has been left out?Show the answer
Everything that follows from a buyer choosing not to order. The three instruments cover loss of goods, failure to pay and movement in a rate. Not one of them responds to a cancellation, a flex-down, a delayed call-off or an expired hold, and those are the events that reach several orders at the same time. The 24.0% is also the wrong measure to be reassured by, because it counts companies rather than causes. The question is what share of the book would move on one piece of news, and no insurer has ever been asked that.
What to take away
Risk on an order book divides four ways: what somebody will insure, what you carry deliberately and are paid for, what you price and refuse, and what you carry because nobody noticed. Only the last is a failure, and it is a failure of paperwork rather than of judgement. Six numbers on one page, limits set from your own record, reviewed the day an order joins the book. And when you fix it, stop before you remove the part you are being paid to carry.