The "Identify one or two simple reasons..." point reminds me of the grand master of turnarounds: Gerstner. When he got the job I called him "the cookie guy" because what the hell did he know about computers (not much). But what he did know was how to turn the supertanker around.
Everybody expected a grand plan and after he did a walkabout and met everybody and learned what the company was doing his decision was: don't change much, just keep at it. (general derision from the press ensued).
He then went and fixed a bunch of widespread problems that fucked EVERYTHING up and so almost everything started to work properly.
Rather amazing.
You brought an interesting point and I have observed it myself too. At many occasions teams would be on the track to improvement but they would crumble because of pressure from above. When a kid can't even walk, regardless of how hard you push it, it can't start running. This applies to engineering teams too.
