Know your customer, not just his money:
the background to Risk Values"

Because terrorism is funded from two fundamentally different sources of money -

1) legally generated and innocently introduced into the financial system; and

2) illegally generated and immorally or illegally introduced into the financial system,

the treatment of the money is divided into two entirely different camps.

When that which is illegally generated and should not be in the financial system moves (or, in some cases, stays where it is), then the established techniques of seeking to identify and, perhaps, contain that money do stand some prospect of success.

But we have to recognise that all commentators on money laundering accept that only a small percentage of the proceeds of crime moving through the world's financial services industry is ever located during the time that it is in an account.

If it is successfully laundered, it disappears into the general economy. Most laundered money that is ever traced and forfeit or confiscated is located because it has been identified after it has left an account and it has been possible to follow a paper trail to an asset which was either immobile or illiquid.

In the case of money intended for terrorism, identifying it after it has left the final account - either by the purchase of a mechanism for terrorism or withdrawal in cash for that purpose - it is likely to be too late.

More, the amounts involved in each individual act of terrorism are small. If a terrorist chooses to use a car bomb, he can - in most developed countries - withdraw sufficient to buy a suitable vehicle out of an ATM in a single (and entirely unsuspicious) transaction. Indeed, purchasing such a low value vehicle with any payment other than cash may be suspicious. So may be the purchase of, for example, box cutters costing a few dollars each. And it is commonplace for travellers to purchase domestic flight tickets with cash - especially tourists or foreign business travellers altering arrangements.

In the case of legally generated and innocently introduced money, the money moves around the financial system without fear of identification as a result of the actions to which it is put - in short, it does what money does in everyday circumstances. As a result, the accepted means of identifying suspicious transactions is probably useless. It is only by formally identifying the account holder as having a connection to terrorism that the account can fall subject to suspicion.

It is clear that the focus on the money is therefore not a valid tool for identifying terrorism funding where that money moves between members of a family or social or other group - because it is doing what innocent money does in the same circumstances. Therefore, the focus has to shift.

There is considerable place, here, for intelligence. But that intelligence has to be accurate and timely. For the financial services business to be able to correctly identify accounts, it has to have precise information as to what to look for. This is information which enforcement agencies have historically been reluctant to publicise - generally for good reason. But a significant factor in relation to many accounts is that names which are translated from any non-Roman script are often spelt inconsistently from person to person and, even, for the same person from time to time. This leads to confusion and - frequently - the freezing of the wrong accounts - leaving the financial institutions exposed to the risk of legal action.

The primary weakness of transaction monitoring systems is that it is necessary to build up norms as a comparator: these norms may be norms for that account (and some users of some systems have privately admitted that it may take three to five years to create an accurate pattern for some account holders) or norms for the bank or for customers of that profile - in this case meaning a customer in that sort of job or business in that sort of area with an account in that sort of bank. Again, the data for these comparators is sketchy - although in several countries at least some of that data is available to governments and forms the basis of Inland Revenue decisions as to which taxpayer to investigate.

Therefore, the only available option is to look not at the money but at the person. Is it possible to examine the motivations of the person when opening an account and, if so, can that person's activity be more closely monitored?

The question of motivation is one which requires analysis of behavioural and personality norms and the obtaining of information from the customer. This information needs to be obtained before there is any financial transaction because, if it is not, then there is a danger that the account will be used for a short time, for apparently normal transactions, but the true purpose of which is to commit offences.

However, in the light of the backlash in the USA against the proposed Know Your Customer laws in late 1999, it is difficult to see how consumers would react if they became aware that financial services businesses were undertaking covert personality analysis on them.

The irony of this is that all manner of businesses do precisely that with the unthinking consent of their customers - it is the basis of the more sophisticated end of the Customer Relationship Management market. Therefore, it is not the principle that would cause the problem but the specific use to which it is put. Financial services businesses would therefore have to be circumspect about disclosing whether such testing is undertaken.

The testing does exist. Developed since 11 September 2001 but based in research over several decades, in response to the all the above circumstances and making a particular point of not relying on any racial, religious or geographical factors - even the applicant's name is not required because of the risk of confusion or prejudice - a testing program has been produced by Risk Values, a joint venture between a counter-money laundering consultancy and a CRM company that produces personality and behavioural profiling for market research purposes.

Turning the usual principles on their head (the usual aim is to produce marketing and products that appeal to the widest possible audience but here the aim is to exclude from heavy monitoring the vast majority of the population who are fundamentally moral) Risk Values analyses the applicant and assigns a Value describing that person's propensity to commit financial crime involving the financial institution - in short to abuse the relationship with the institution.

Risk Values does not claim that its product can prevent terrorism or financial crime - but it can provide a warning about a particular customer when no other approach works.