An interesting paradox exists for major programmes: more and bigger programmes are being planned and built despite their poor performance record in terms of cost overruns, schedule delays, and benefit shortfalls.
For the majority of programmes, performance is significantly and consistently below what could be called "best practice " – or just "good practice" – when measured in these terms. This has been the case for decades and existing data show no immediate end to this state of affairs.
For more about the major programme paradox, see:
Megaprojects and Risk, Google, Bing.