Compliance rules could be extended to include due diligence on partners, and block contracts and payments to and from entities that have no clear ownership.
It could but realistically it's unlikely to happen. Nobody would ever accept taking responsibility for dozens, hundreds, or thousands of partners, and the expenses involved in validating every last one of them. Not to mention the lost business on either side. Even the IRS or banks can't keep up with long ownership chains or properly identifying customers and they have a more vested interest.
Laws and rules are far slower to adapt than the workarounds that bend them. And clear ownership says little. Everything can simply point to a more or less real identity that nobody will ever find.
Don't get me wrong, I'm not against such rules. I'm just saying that the likelihood of them achieving the results we imagine are slim.
How hard is a policy stating maximum of n-levels depth of nested ownership allowed and any kind of cyclical ownership causes loss of control in both ways?
It's hard because non-fraudulent corporations commonly have very complicated ownership structures. The global corporation owns the US subsidiary which owns the Auto subsidiary which owns Auto Parts subsidiary which owns the Northeast Region Auto Parts subsidiary which owns a subsidiary that makes transmissions which owns a subsidiary that makes warranties for transmissions etc. etc. The level of complexity necessary to confound an external investigation is not inherently less than the level of complexity that exists in legitimate real corporations.
Corporate ownership is also traded as a commodity. The US Auto Parts Northeast Region Transmissions Warranty subsidiary is required to hold assets sufficient to bond the warranties it issues, so instead of keeping a large pile of hundred dollar bills on the coffee table, it holds shares of the parent company and you now have circular ownership. If the parent company is loaded up with debt then you might find that the subsidiaries like that own the majority or entirety of the parent company -- without even necessarily realizing it.
You also have the problem that people willing to commit fraud are willing to commit fraud. Convince some credulous victim that all they have to do is put in $1000 and sign some papers and they can invest in The Next Big Thing and they've not only scammed them out of $1000, now it's the victim's name on the shell corp.
Generally speaking identifying people who don't want to be identified is hard, the methods effective to do so oppress innocent people more than anything, and it is best to look to other solutions. For example, if you discover people are publishing misinformation, show the truth to the same people who saw the misinformation.
I'm just talking about the expectations. Writing the rule is not the difficult part. Enforcing them is because it's extra effort and expense and it leads to losing some business. Shareholders are more than happy to support a company's ethics until they stand in the way of profit.
Finding an appropriate n might be a delicate, many big (legitimate) businesses have really complicated ownership structures. Then you have to task someone with validating all of this. Whose expense is it to do the due diligence and check everything? WHo takes responsibility for failures in this regard? And finally it will mean some business is lost on both sides and it might be a lot of money.
Banks have anti money laundering policies yet still find ways to go around them all the time because they want the business. Banking secrecy wasn't there to protect privacy but to obscure shady activities. As I said, it's good to have the rule and I'm not against it but I have the feeling people have some unrealistic expectations from this. They just raise the bar a bit and filter out the "chaff" while still allowing larger interests to prevail.
> Finding an appropriate n might be a delicate, many big (legitimate) businesses have really complicated ownership structures
It is something between 10 and 13, if you accept that governance structure of an organisation must match its social communication structure. It is something between 14 and 16, if you accept that governance structure of an organisation must match its business divisions. Any n>16 is prone to communication shortcutting.