I see that attitude working with individuals. Most people takes their responsibilities seriously.
But for corporations it would never work. Corporations would do literally nothing and take the money if there were no consequences. Even when all the employees are good caring people, the corporation incentive is to maximize profit. Corporations are less than the sum of its parts when it comes to morality.
I think this way of thinking is crucial. Ultimately, these projects start to take this shape at conception... long before the failures are realized. The heads-i-win-tails-you lose games. The liability avoidance. The negotiation of guilt for the expected failures down the road.
It's a similar game in the corporate world, like with enterprise software. Everyone knows that an ERP project will blow the budget and timeline. Lots of lawyerly hours are spent creating liability frameworks for it. The result is always ambiguous, pinning failures on specification failures, changing requirements or whatnot.
Systems respond to incentives in many ways, but institutional (government, commercial, NGO, charity...) response to liability is rarely a positive sum game.
Not saying I have the answers, just that the obvious ones have been tried and failed.
This isn't a marginal issue. As it stands now, there is are huge categories if things we need and can't have... because the systems for getting them are horribly ineffective.
If trains cost 5X what they could cost, we effectively can't have new public transport systems. That's not sustainable.
This is how BAA built Heathrow Terminal 5. They took on all risk and just dealt with each problem as they came up: https://www.christianwolmar.co.uk/2005/04/project-management...
[1] If you take projects with a 10% chance of bankrupting you, you'll be out of business in no time.
The article was about BAA ( British Airports Authority originally) building the physical terminal building with a roof and ceilings and power and networks. Different organisation and set of project managers.
Still shows that public memory of projects which go well in one area can be effected when someone else screws up.
Not that this would every happen in software projects.... ;-)
Even in aviation, however, independent investigations by government (e.g. the NTSB) do not extinguish civil or criminal liability. Incident investigations are inadmissible in court, but civil suits and/or criminal prosecutions can be, and are, conducted as parallel processes.
Clearly this is because they're being sued, and they want to minimize their losses. Sometimes I wonder what the world would be like if we just said "that sucks" and required a full postmortem instead -- no legal liability, just a thorough public review that everyone could read and learn from. We see this in other fields, like aviation; the NTSB investigates and publishes their finding of facts. The field learns from the mistakes, and as a result it's incredibly safe. (Even if you account for random chance, like MH370 or the whole 787MAX debacle, we're still doing better than we were 10 years ago.)
Instead we just keep making the same mistakes again and again. Everyone pretty much expects to get on a plane and arrive at their destination alive. Everyone also expects their megaproject to be delayed and be finished over budget. When are we going to fix this?