Jeez, can we stop with the meaningless arguing? Intel has had a huge lead in CPU perf for years now, and no, it's not a coincidence. At the same time, CPU performance hasn't increased much during this time either. Has their lead led them to be slightly lazy? Possibly. It's impossible to speculate whether this is true or not. All we know for sure is that Intel has been more focused for the last four generations or so on reducing power consumption than increasing raw power. And, we should add, this has caused them to more or less own the entire laptop market (the biggest consumer PC market), while keeping their lead in desktops and dominating the server market. Their architectures scale remarkably well (4-165W? That's a huge range!).
Is the reason for thinning the Skylake PCB cost cutting? Possibly. It's hard to argue that it's not at least part of the logic behind the decision. Does it matter? No. As we've seen, with ~99% of all cooling solutions, it makes no difference. I'd bet the thinking behind it is along the lines of "Can we make the PCB thinner without any negative consequences? *Testing* Sure, seems like it. Let's do it!" After all, there is absolutely no logic in wasting materials, no matter the scale.
Did they screw this up? Possibly. Should they have tested this more thoroughly? Sure. Has it yet affected any end users, at all? Not that I've heard of. That'd be a class action suit waiting to happen.
The use of a subpar TIM is another matter entirely. But still one that doesn't matter to 90%+ of users.