Was that because the experts in their fields underestimated from the beginning, or because the programme managers pushed them into corners to give unrealistic timescales and budgets, as, 'The customer won't like the original figures'?
All of the above.
In my experience, the team which delivers the forecast is not the team who has to deliver the project. Also, I find that we engineers are rubbish at estimating how long tasks take, usually wildly under estimating.
Recently, I have been involved with a 'lessons learned' for a non-aerospace military project I was involved with that went completely pear shaped and the main points were total underestimation of timescales and costs and that all the bad decisions were taken right at the start.