Findings: - ~4% (20M of 500M companies) have shell-company indicators - Indicators are limited to patently obvious: most-commonly bulk registrations and "atyp

Findings: - ~4% (20M of 500M companies) have shell-company indicators - Indicators are limited to patently obvious: most-commonly bulk registrations and "atypical directorships" - because it's easy to register (in the UK), but hard to make up people (hence getting pseudo-directors) - Most are in the US (unclear if percentage-shell is higher) Reflections: - 500M companies for 6B people is ~1 company per 12 humans. Very high! - Post-Panama-Papers drop in prevalence is likely just improvement in shell quality to avoid obvious indicators - Still, 4% or even 10% could be manageable, but effective management would drive a jurisdictional race to the bottom - Unclear to me: are private data brokers ahead of the government in making actionable analysis available to investors? Has anyone (in academia) come up with an ownership/governance model that's traceable but privacy-preserving for different applications (market monitoring and investment due-diligence; criminal, civil, or tax investigations)?