Thanks Sunny.
How do we estimate the number of uninsured or underinsured deaths/injuries/hull losses/dings that would have occurred, but for flight service?
If we can do that, we can then multiply it by (e.g) $4,000,000 or whatever the going rate is for a death, and the same for the estimated number of injuries/hull losses/dings, to work out whether flight service is ‘worth’ what it costs.
Presumably that's what AA should be doing.