That's a good point. I think of any given project as a CDO. There are good, productive debts and bad, stinking, potentially crushing debts in there. What matters is not that there is debt - I think there will always be - but what the blend is.

If the debts were thoughtful and deliberate with understood risks, they might make perfect sense.

If the debts were taken by people who couldn't evaluate them and had no business making that decision, it's going to be bad.

Odds are the first are well-understood and maybe even documented. The second may explode when the underlying assumptions or risk changes.

Ref: https://www.investopedia.com/terms/c/cdo.asp

or if you'll ever have to pay it back. Just because you have some tech debt doesn't mean you'll actually have to ever deal with it.

Or the unit of measure. It's really hard to tell what sort of technical debt you are taking on and building for contingencies that may never manifest is it's own type of sunk cost. That's (a) why projects need to have some strategies that move forward without excluding lateral (or minor-backtracking) moves, and (b) why software development is inherently hard.

It's more like a compound interest. A financial example would be payday loans with an incredible APR and a low monthly payment, your debt is forever increasing.