I think there's also an element here of... when you work a long time on something, you tend to become emotionally attached to it, and you want it to be good and work well. Time to fix technical debt and work on making the product good is often implicitly waved as a carrot ahead of people who care about that sort of thing. "We can prioritize that after we ship $FEATURE."

This then feels like a betrayal on an emotional level when instead of a nice block of time to fix technical debt, instead the priority becomes $NEXT_FEATURE (the "features that don't exist yet.")

Management that can successfully keep the treadmill running ships features faster, so it keeps happening, and can contribute to burnout as (what felt like) implicit promises are repeatedly broken for the good of the business at the expense of the product.