Randy (funny gamedev guy from YouTube) said in a recent video that he realized he'd spent the last ten years making engines to avoid the creative terror of action making a game. I'm paraphrasing slightly, but that's what it came down to.

"I thought if I made a really good engine, making a game would be the easy part!" I had similar thoughts when I was younger. Surely if I just upgrade my tools, the hard part will become the easy part!

Jonathan Blow says making engines is easy, because enginedev only takes a relatively small part of development — the game itself takes way more time and energy.

So his argument is, in the grand scheme of things, the engine is not that much work. (Since you're gonna spend ten years working on the game anyway, of course ;)