Most people in Hong Kong accept the idea that banks pool credit information about consumers.
By sharing details about non-payment of debts or bankruptcies, the banks can avoid lending to people who are bad risks — especially those deliberately planning to default.
That benefits not only the banks but consumers with good credit records.
The insurance industry in Hong Kong would like a similar sort of system to help fight insurance fraud. We lag behind many other markets in this area.
There are various types of insurance fraud. For example, someone making a motor vehicle insurance claim can exaggerate the costs of vehicle repair after an accident or fake an injury.
In some cases, the deception can be quite complex, for example when syndicates are involved.
Needless to say it is a crime — and law-abiding insurance customers end up paying for it through higher premiums.
At the moment in Hong Kong, insurers can investigate suspected fraud on a case-by-case basis, or they can use external agencies to do such work.
Only a small number of insurers have dedicated fraud investigation units and only a minority have any sort of automated procedure to alert officers of possible fraud.
None are thought to use sophisticated data analysis to predict fraud in Hong Kong.
Past efforts to address this issue have not been very successful, despite hard work by many people in the industry, including insurance functional constituency representative KP Chan.
Many insurers believe fraud is going undetected in Hong Kong. And most believe that the ideal solution is a centralized insurance fraud database, in which participants could pool information about fraudsters or people suspected of fraud.
One of the biggest hurdles is the issue of personal privacy. However, the experience of many developed economies shows that there are ways to safeguard privacy while strengthening action against fraud.
The consultation has involved interviews with a wide range of insurers, plus the Hong Kong Police commercial crimes bureau and the ICAC.
Work is now continuing on gathering industry feedback and reaching out to the Consumer Council and regulators.
If the plan is accepted, the database will initially cover medical, accident and medical insurance, and subsequently be extended to include employee compensation, travel and life insurance.
The data within it will be specifically relevant to claims fraud detection.
Only specific claims handlers and investigators will be able to access it. Underwriters — the insurance company departments that process applications for insurance policies — will not be able to access the database.
As with banks, insurance customers will be fully notified about what sort of data may be collected before they take out policies.
Privacy considerations will also apply to information on third parties. For example, information on garages will not be a problem as it does not involve personal data. Information on medical practitioners will be more limited. As with all personal data, there will be strict rules on data retention and fair treatment.
In the US, UK, Canada and Australia, the insurance industry estimates that about 8 percent to 15 percent of motor claims payouts are fraudulent.
Fraud varies according to type of insurance and other factors. By definition, it is hard to detect as the evidence is not normally visible to insurance companies and law enforcement.
The industry and consultants here in Hong Kong believe that fraud here is mostly not very sophisticated and probably not as frequent as in some other markets. However even taking a very conservative assumption — say a 5 percent rate of fraud in claims payouts — the proposed database would pay for itself.
Essentially, insurance fraud is a sort of theft and ultimately the whole community ends up paying for it in some way.
The evidence from this project suggests that a centralized insurance claims database in Hong Kong would benefit insurers and honest insurance consumers.
– Contact us at [email protected]