> Take the number of vehicles in the field, A, multiply by the probable rate of failure, B, multiply by the average out-of-court settlement, C. A times B times C equals X. If X is less than the cost of a recall, we don't do one.

Assuming that Sig Sauer management is reasonable, we can assume that one or more of these are true:

* The known rate of failure B is determined to be low. Consider that not every discharge would be from a design flaw. Many cases can be assumed or proven to be user negligence.

* They assume that they can keep the court settlement costs, C, to a low value by never admitting fault and hoping that no one else can convincingly demonstrate a poor design. Many cases result in no injury or non-lethal injury, which naturally reduces C.

* The number of guns produced, A, is quite large, so the cost of the recall is also quite large.

* The unit cost of the recall (X/A) is much higher than known externally. This is my preferred theory (outside of corporate incompetence & malice). It could be the case that the design has an issue with tolerance stacking AND there is no single dimension of replacement part that resolves the issue. You could imagine that the replacement part needs to take up negative tolerance by being slightly larger, and positive tolerance by being slightly smaller. Without carefully measuring each unit (which is expensive), you can't determine which part to use. Or it could be that the part that would need to be replaced is a substantial part of the weapon's cost, e.g. the slide or the frame.