Defense business accounting does not help
In *most* industries a new product is priced roughly by taking the startup costs and production costs and working out what price customers will *likely* pay for the goods. In reality it's likely in several price points will be worked out and from those a "break even" number of sales worked out. *If* product sales > break even -> Profit on *any* further sales (and the element in the costing to cover start-up is now *also* profit). If not tough.
In defense.
There is typically *one* customer. (Pop quiz. How much of UK designed stuff, mostly from BAe, found *other* foreign customers *outside* the original customer or partnership?)
When you've *finally* got a workable design (AFAIK still a pretty profitable process as I suspect Cost+ is still pretty common in the UK defense biz, although *perhaps* not quite as common as it was).
Tot up *all* the start-up, mfg costs and divide by the number of unit the MoD is asking for.
*That* is the price they pay. It also the reason why the next (if there *is* a next) batch ordered in a reasonable time frame (IE before they lay off the workforce and break up the tooling) is substantially cheaper.
It's a subtle difference but quite an important one. It makes the defense biz about a^&e backwards to just about *every* other product in the world.
Mine will be the one with some spare Ferranti f100L's in the pocket. The world's first (and only) bipolar single chip microprocessor sold AFAIK to 1 customer.
Care to guess who?