Why isn't there a centralized black list of addresses that are known to have received fraudulent orders? Just like with IP addresses that are known to have spammed.
First of all, you don't block the order if billing and shipping addresses are the same, even if blacklisted. Another important thing would be, an address is blacklisted only in the case of a proven fraud, not just suspected. Next, apartment/house sharing would be problematic for the tenants, well then co-habitants or the landlord will have a good reason to throw the fraudster out.
But how would the landlord or co-inhabitants know? And even if it were proven, it would still not be a straightforward task to deal with such people.
The aggressive self-righteousness described in the article is very typical of criminal types. This makes them very convincing at playing the victim when they are challenged. They can very easily turn the tables on you, even to the point where you are the one facing sanctions, and even paying compensation.
However, they're usually curated and distributed for enterprise customers, which means you may be paying a huge subscription fee (6 figures yearly wouldn't be uncommon) for access to that list.
Fraud is such a big problem that companies can easily charge a massive premium for anti-fraud services and software.
They're not perfect though, so you get some false-positives and they'll let through some e.g. Spammers sign up for web hosting in my experience.