Whether to insure this shipment at all

Do the sum properly, including the part that argues against you

A course on cargo insurance that ends by saying always insure has earned nothing. So here is the arithmetic, done against Thistlewood's own record rather than an industry claim.

Thistlewood move 46 import containers a year, averaging USD 167,200.00 of insured value. That is USD 7,691,200.00 declared to Vasteral in a year. The rate on the open cover is 0.075%, so the premium is USD 5,768.40. The excess is USD 1,500.00 on each and every claim.

Their claims history over five years: 3 claims, averaging USD 10,900.00 gross. That is 0.6 claims a year, and USD 6,540.00 a year of expected damage.

Insuring costs USD 128.40 a year more than not insuring. That is the honest number, and it is very nearly a coin toss. Which is exactly what you would expect: an underwriter charging much less than the losses would not stay in business, and one charging much more would not keep the account.

That number answers the wrong question

Look at what is not in those five years. No total loss. No general average. No salvage. The record is three ordinary partial losses, and it is being used to price the two events that would actually matter.

Put those two events into years of premium, and the whole question changes shape.

Twenty-nine point nine years of premium, arriving in one week. That is the product being bought. Insurance is not a good bet and was never meant to be one. It is a way of turning a number you cannot carry into a small, predictable one you can budget for. The USD 128.40 a year is not the cost of the cover. It is the price of certainty, and put that way the decision usually makes itself.

What else is inside the premium

Four things arrive with the policy, and none of them shows up in a comparison of premium against expected damage.

• The general average guarantee (lesson 6), issued free, in two days, instead of a cash deposit on a five-day deadline.
• The recovery (lesson 5). Vasteral chased Nordvahl, at their own cost, in Thistlewood's name, for eighteen months. No merchandising team runs that.
• The surveyor at the port, appointed with a telephone call, through a network your company could not build.
• The certificate a bank will accept. Where the shipment is paid under a documentary credit, the insurance document is one of the papers the bank checks, and can refuse. That belongs to track 13, which owns the credit and its discrepancies. It is another reason the cover is not optional in practice, even where it looks optional in the arithmetic.

The decision that IS worth arguing: how much you keep

Whether to insure is usually settled by the tail. How much of each claim you keep is a real calculation with a real answer, and it is the one most companies never do.

Vasteral offered Thistlewood a lower rate against a higher excess: 0.052% if the excess rises from USD 1,500.00 to USD 7,500.00.

The higher excess is USD 1,831.02 a year worse. But notice the shape of the answer rather than the answer itself. The saving is fixed, and the extra you keep grows with how often you claim. So there is a frequency below which the higher excess wins. It is 0.29 claims a year, a shade under one claim every three years. Thistlewood run at twice that, so they should stay where they are. A company shipping the same value with one claim every four years should take the higher excess and pocket the difference.

Do that calculation with your own frequency, before your broker does it with a market average.

The one case where not insuring is right

There is one, and it is about the minimum premium, not about risk.

Vasteral charge a minimum of USD 45.00 per certificate. On a full container that does not matter: the rate premium on TW-8807 was USD 129.34. But on a USD 900.00 sample parcel to a buyer's technical team, the rate would produce well under a dollar, and the minimum charges USD 45.00. That is 5.0% of the value of the parcel.

At that price you are only buying cover if you think the parcel has better than a one-in-twenty chance of never arriving. It does not. Send it, absorb the loss if it happens, and spend the money on the containers.

That is the whole exception. It is about a fixed charge swamping a small exposure, not about a judgement that goods are safe.

What one bad shipment actually cost

Put the whole course on one page. This is TW-8807, from the trailer on the port road to the adjustment two and a half years later.

The policy paid 244.4 times its own premium on this shipment, which proves nothing at all. One shipment never does, and the annual table at the top of this lesson is the honest test. What the second table proves is the thing worth leaving with: the insurable half of a bad shipment is about half of it. Everything in lessons 2, 3 and 4 — the uninsured leg, the delay exclusion and the packing exclusion — was decided long before the weather, by a purchase order, a certificate and a packing specification somebody could have read.