The problem with the "money" statements is they don't provide enough information to track veracity.
For example, the claim GM was losing $10K per vehicle. Was that amortized over the entire production (to include all the front end research, development, tooling, advertisement, manufacturing, warranty)? Or, is that number strictly a production loss. Literally losing money not even considering all the sunk costs?
Was it a loss figure for the loss rate (amortized or production) at the moment in time, or averaged across the entire development or production.
And what of the salaried work force. Some of those engineers worked very long hours on the kappa. Without the kappa, maybe they would have worked less hours...yet the pay would have been the same. Are those R&D dollars, that would have been spent anyway, levied against the kappa?
Knowing how the accounting is performed is very important. GM might be losing $10K per vehicle amortized over the entire development effort, but might be making money during production. Big difference!
And one also must consider the peripheral benefits, such as employment of a work force that would otherwise be unemployed. The revenue generated at the dealers....both sales and service. And the generic halo effect. Granted, these peripheral benefits don't necessarily map directly to the bottom line, but they affect that bottom line in indirect ways.
I bought a SKY. I get my service performed at a GM dealer...who buys OEM parts from GM. So GM is making money off the sale seven years later. I'm pretty sure that wasn't factored into the loss statement.
Either way, the kappa volume was too low to have much of an impact either way from a money standpoint. Even if the program was a wild success and made big time profits, the volume was too low to impact the disaster that unfolded. By the same token, even with the loss, it was only a footnote on the balance sheet flowing red.