I suspect this is a system where the result is inevitable:
- Positive feedback: earning money for your company.
- Negative feedback: none
- Resources: vulnerable people
Adding negative feedback (eg watchdogs, regulators, financial consequences) can get you most of the way there to fixing this, but even a few remaining % of abuse is still a lot of actual vulnerable people getting mistreated. A well engineered solution to this problem would include removing the positive feedback; not just adding negative feedback.
EDIT: Addendum: Unlike financial resources, which can be fixed/repaid/etc when something goes wrong in a few % of cases, vulnerable people can't always be "fixed" after abuse. This is a key difference that the people making these policies and systems seem to be unaware of.