I've called the phenomenon of private corporations refusing service the "Maoists in the Risk Department" in the past.

The reason why risk departments all inevitably reinvent Maoism is because the only effective enforcement mechanism they have is to refuse service. Fraudsters are fundamentally illegible to businesses of this size. And as the article stated, recidivism rates in fraud are high enough that someone caught doing fraud should never be given the time of day ever again. So the easiest strategy is to pick some heuristics that catch recidivist fraudsters and keep them a jealously guarded secret.

This calculus falls apart for the government. If someone rips the government off, they can arrest them, compel the production of documents from every third party they've interacted with, and throw them in jail where they won't be able to rip anyone else off for decades. Obviously, if we gave the Risk Department Maoists these same permissions, we'd be living under tyranny.

Well, more tyranny than we already live under.

But at the same time, the fact that we have these legal powers makes Risk Maoism largely obsolete. We don't need to repeatedly reinvestigate the same people for the same crime "just in case".

My take is that we lack granular punishments.

Right now we have either some form of fine, and while this can be incredibly painful, usually is not, then we go straight to like multiyear prison sentences, with perhaps a few suspended sentences in between there.

I dunno, maybe a world where "you did a small bank fraud so now you have some kind of antifraud system attached to you" is genuinely worse than the one we live in now, but the idea of being able to target more specific aspects of someone rather than just prison/not prison seems interesting.

I guess we have stuff like "not allowed to use a computer for 5 years" (thanks hackers movie!), dunno how effective or practical that is.