Building up a profile of a fraudster

Greg Au-Yeung surveys the strides in risk management that help companies limit the damage

PUBLISHED : Monday, 15 October, 2012, 12:00am
UPDATED : Monday, 15 October, 2012, 1:53am

Some of you may remember the sci-fi movie Minority Report, starring Tom Cruise as a police officer in the year 2054 who apprehends criminals based on foreknowledge provided by psychics - that is, he arrests potential criminals before they have committed a crime. This may be wishful thinking. Or is it?

According to the 2010 Association of Certified Fraud Examiners report on occupational fraud and abuse, a typical organisation loses 5 per cent of its annual revenue to fraud, with a median loss of US$160,000.

Risk management plays a crucial role in the consumer finance business.

In China, both personal and business credit information is kept in the government's credit bureau, under the People's Bank of China. The service has been live since 2006; an individual's personal and debt information is uploaded and updated regularly by financial institutions.

At the end of last year, eight billion personal records were already registered, with more than one billion credit histories. Since 2011, the bank has issued new rules to open up its database service to financial institutions other than banks, such as microcredit and guarantee companies.

Many people are unaware that, according to Chinese law, failure to make credit card payments is a serious crime. "Delinquent payment", as it is known, is considered a "malicious overdraft", which is subject to a maximum jail term of 10 years and a fine of 500,000 yuan (HK$613,000). Hence, people should think twice before committing such crimes.

As a first line of defence, there are various anti-fraud methods, from simple phone-number and address checks, plus blacklist validation, to sophisticated artificial intelligence that can identify potentially fraudulent loan applications.

For example, the probability of a particular person defaulting can be predicted. Another approach is to look at which demographics or market segments to avoid to reduce risk.

Fraud is not only committed by dishonest borrowers, but also by employees; there may be collusion with outsiders to obtain a loan, for example. A 2011 global survey of fraud by the risk consultant Kroll found that 60 per cent of fraud was committed by employees, of whom about half held senior positions.

Many local Chinese financial organisations operate with a small team, with no proper structure in place such as an internal audit department to perform regular assessments, an information security policy that prevents leaks of confidential data or a platform to monitor, assess and respond to unscrupulous behaviour.

Risk management discipline was introduced in China's fund management industry just under 10 years ago, and has been widely accepted as best practice.

Unlike banks, fund houses do not employ vast numbers of staff.

Microcredit companies need to rethink their risk management policy for their own survival in the long run.

The government also needs to provide more support - in terms of favourable policies and incentives - for the microfinance industry to flourish.

The best scenario is to identify the bad guys in the first place, so you don't lend them money. Unfortunately, there are always some unknowns that cause problems, such as an inefficient sales team, shrewd borrowers or a bad credit policy.

This is when a money-recovery strategy comes into play. This is more art than science, and credit collectors often need to use their wits and perseverance - the personal touch - to get the money back rather than rely on statistical analysis.

ZestFinance, an online- lending alternative start-up founded by former Google chief information officer Douglas Merrill, has recently filed a patent application on its credit-decision model that optimises credit risk analysis. It allows multiple underwriting models to run in parallel, which aims to extend credit to 25per cent more borrowers and yields a 20 per cent increase in repayments.

Some banks are already building data repositories that tap into social media, hoping that one day this information may be useful; there may be indications of a person's financial status and personal changes, as well as behavioural trends from his or her particular social circle.

We must accept that we live in an imperfect world with corrupt data and dishonest human beings. However, fraud can be avoided, just as risks can be mitigated, even in China. The industry is progressing every day, finding better ways to identify, detect and prevent risks, with the aid of technology.

Nevertheless, it is still down to human beings to make the right policies, as well as business and governance choices. Bloggers should perhaps ponder what they write, in case one day their mortgage application is rejected because of what they said in cyberspace.

Thankfully, thoughts remain private - at least for now.

Greg Au-Yeung is chief information officer of VantAsia (formerly AIG) Finance in China and a visiting professor at Fudan University.