There are as many fraud schemes as there are fraudsters, but they fall into a few general buckets. Here are two of the most common fraud scenarios and solutions for solving them.
A credit card issuer experiences fraud losses when fraudsters use fabricated account information, usually with a real SSN.
A card issuer offers credit cards via an online channel. Taking advantage of a potentially loose credit policy, fraudsters fabricate bogus identities. In most cases, the fraudsters concoct names and addresses, and give each identity a birth date within the 21 to 23 age range. Then they assign real SSNs to each fictitious applicant.
To uncover the synthetic identity, screen each identity component. The crucial first step in recognizing and combating synthetic identity fraud is to scrutinize the identity’s components (such as name, address, Social Security number and telephone number). Are they accurate? What is known from existing records about portions of the identity? The overall goal is to verify the existence of this identity.
Criminals use stolen identities to open new accounts via a lender’s online or call center channel. It is fairly easy for a criminal to obtain personal information (such as name, address, SSN and date of birth) to use in a fraudulent manner. As a result, identity theft continues to be a major white-collar crime in the U.S. Identity theft can be defined as any act in which someone uses the personal information of another without that person’s knowledge or consent. A common term for this type of identity theft, in which the fraudster poses as the actual consumer, is true-name fraud.
Relying on the anonymity of a lender’s online services, a fraudster uses stolen personal information to apply for an account in the victim’s name. To detect attempts at true-name fraud, especially online, organizations need more than simple fraud tools that cross-check applicant-supplied information with various databases. Anti-fraud tools that just check the validity of the name, date of birth and address, for example, don’t thwart criminals who have already obtained that information through some means of identity theft.
Use a combination of authentication and modeling techniques. Studies by Equifax show that predictive tools based on user behavior, velocity of activity, and known fraudulent behavior are an effective aid in spotting fraudulent applications in a real-time environment. Predictive modeling can demonstrably help reduce losses from true-name identity theft. Optimum performance is achieved when the model draws upon a variety of data types, including:
- A business’s own data identifying location, time and channel of past fraudulent incidents
- Credit data for the consumer
- Collections data for the consumer
- Other proprietary information, such as fraud consortium data