Predictive money laundering and damming orders – what are they?

By Christine Duhaime | July 10th, 2017


Last week, the US Treasury Department imposed sanctions against a bank in China for allegedly providing financial services to North Korean companies and said it was rolling out more sanctions in the weeks to come.

The bank, the Bank of Dandong, is alleged to have facilitated money laundering by allowing its bank to be used to move money connected indirectly with WMD. Selling, trading, financing or providing services in support of WMD is a predicate offense for money laundering and the proceeds of the offence is proceeds of crime, hence the money laundering allegation against the China Bank of Dandong which, if the allegations have merit, profited off the financial transactions and its revenues therefrom are proceeds of crime.

The bank is alleged to have provided banking services for the sale of coal, not directly for the sale of WMD. But North Korea is believed to use its proceeds from coal exports to fund its WMD program.

In May, the US DOJ had obtained what is called a “damming seizure” warrant.

A damming seizure is an unusual court order because it is granted on the basis of the predictability of criminal activity in the future and it covers the full scope of financial transactions going through a bank of the targets.

In essence, it forces banks to disclose all incoming financial transactions of the target and to block outgoing transactions, including correspondent banking transactions. The damming order was not for funds held in the US — it was for funds held offshore.

In that sense it is similar to a mareava injunction only with a wider scope and full disclosure. The constitutionality of an order based on the propensity of criminality is interesting and was litigated in the US. A damming order is anticipatory – it is not based on existing probable cause – rather it is based on cause existing at execution, namely that funds are going to be laundered at the eight banks subject to the damming order and therefore the order should be installed in advance.

The US courts held, among other things, that damming orders based on future criminality were constitutional in the US based on 2006 precedent from the US Supreme Court.

Eight banks

The damming order was served on eight US banks for funds held in the US and elsewhere but that are tied to the US by virtue of the correspondent banking relationship. The US alleges that the eight banks processed over US$700 million in financial transactions that were illegal because they were sanctions under US sanctions law.

The eight banks are:

  1. Bank of America
  2. Bank of New York Mellon
  3. Citigroup
  4. Deutsche Bank
  5. HSBC
  6. JP Morgan Chase
  7. Standard Chartered
  8. Wells Fargo

Damming orders are not new – they are usually used for transnational criminal organizations, such as the Mafia, to monitor their financial transactions between countries to learn with whom they associate and where the movement of illicit goods and money is going to and from.

Predictive AI used to guess future criminals

Interestingly, there is a debate in financial crime and artificial intelligence circles about the use of predictive AI that predicts the propensity of criminality of a person based on, inter alia, their family, business and social media connections.

In financial crime, usually a person associated with organized crime will “infect” those people around him or her, and eventually a non-criminal will engage in criminal conduct, statistically speaking. What predictive AI does is identity those persons and in some pilot cases, analysts then enter such persons in a database for monitoring by law enforcement to protect society. Theoretically, this may violate the presumption of innocence and other rights but based on case law associated with this case’s damming orders, it would appear that in the US, courts have determined that the greater interest in protecting society against the dangers of financial crime outweigh certain personal potential rights violations of persons or companies. In a round-about-way, this case could be seen as a win for the use of predictive AI by law enforcement.

Comments are closed.