I think it's more like they don't want to do it. If they consider someone a "fraud", why allow this "bad actor" to be still in the system? It's like "if someone steals, put them in the jail altogether, instead of just ban him from grocery stores".
It still seems overly harsh at the personal/individual level. Should lose your photos because you make a ghastly comment on YouTube, or lose your email access because you app. got pulled? A Google account has incredible breadth. I understand they have to deal with an incredible scale of fraud but surely they can differentiate between a recently created account using a few of their products for fraud vs. an account that has human-looking activity over a longer period of time.
I totally agree of differentiating punishment from behaviors. A bad comment shouldn't cause all the pics from Google Photos to be inaccessible. But like you said, Google probably has to deal with these issues in incredible scale, so unless some high up personally deal with the individual cases, almost all of the bans are automated.
Also, considering all the intervention would incur some cost, especially with high ups involved, unless unbanning the account has greater benefit to Google, why would they waste time doing it?
