If missing estimates of time and resources is a success, then what would be a failure? Just giving up? Can we make any project a "success" by just keeping on throwing time and money at it until it's finished?

People (like executives, or voters) made the decision to go forward with these projects based on the estimates. If engineering estimates can't even be accurate to within a factor of 2 (like these), then what's the purpose of providing them? If humans are just bad at estimating, why are these estimates always much lower than actual, and never higher? What about the benefits which could have been achieved by other uses of this money? How can anyone make an informed decision about which possible engineering project to implement, when estimates are this bad?

These days, I assume any estimate without a confidence interval is a lie.