One of the reasons to prefer SQLite even in simple cases is that "filesystems are hard" and "transactionally written back to the backing csv" has a ridiculous number of failure modes, many of which SQLite handles correctly: https://danluu.com/file-consistency/

But in many cases, yeah, simple file use is good enough, as long as important stuff is backed up somewhere / a human can easily re-upload and repair anything needed. It's a <0.1% optimization, it really only saves you noticeable effort when you're doing like millions of those operations per day.