Magazine article Risk Management

Insurers Tackle Fraud with Technology. (Fighting Fraud)

Magazine article Risk Management

Insurers Tackle Fraud with Technology. (Fighting Fraud)

Article excerpt

Virtually every enterprise is affected by losses due to fraud, but industry segments that process large numbers of transactions are particularly vulnerable. Every day, the number of electronic transactions continues to grow, increasing both the potential risk of fraud and the potential size of fraud losses. Analyzing the large volume of data from these transactions to detect fraud and abuse is simply beyond the ability of human beings, however skillful they may be. So to keep up with the times, insurers are increasingly turning to neural network technology. Neural network fraud-detection systems mark a significant jump ahead in the insurance industry's technological race against fraudsters, and may save insurers--and their clients--millions of dollars.

How Fraud Got Out of Control

Insurance fraud costs the average American household more than $5,000 a year in the form of higher premiums and prices for goods and services, according to Insurance Fraud: Renewing the Crusade, a recent study from Hartford, Connecticut-based Conning & Company. In 1999, consumers paid an estimated $96.2 billion in increased premiums and more than $530 million in the increased cost of goods and services--all thanks to fraud. The epidemic stems from various sources:

1. Disconnected claims processes. The paper-based nature of the claims payment process, the disjointed communication structure, the lack of division between fraud screening and adjusting duties, and workflow inefficiencies allow insurance fraud to go undetected.

2. Inefficient fraud detection. Fraud detection relies mainly on an adjuster's ability to spot, amid thousands of reviewed claims, a pattern of treatment or billing that indicates wrongdoing. Fraud is also identified by anonymous whistle-blowers who call toll-free numbers to report suspicious activities--another highly circumstantial and inefficient method.

3. Lack of investigative resources. Investigators are swamped with too many cases and frequently waste time on false leads. Cases are often referred to them too late, if at all. Consider that the entire domestic property/casualty insurance industry had 116 million claims reported in 1999, according to Deutsche Bank. That same year, there were only 10,000 special investigation unit actions.

4. The economy. Most insurance fraud is opportunistic. People see a chance to make money from an insurance claim, and they take it. For example, layoffs due to a recession, coupled with record levels of personal debt, can prompt downsized workers to look at workers' compensation claims as a way to make additional money. Eighty-four percent of respondents to the Conning & Company survey say that insurance fraud will increase if the economy worsens.

5. Organized fraud schemes. The Washington, D.C.-based Coalition Against Insurance Fraud (CALF) says that organized crime plays an increasingly pivotal role in insurance fraud. Because insurance and health care fraud is relatively easy to perpetrate, organized crime rings recognize it as a huge opportunity.

6. The perception of a victimless crime. Although 88 percent of Americans recognize that fraud is a major force behind premium increases, one in three people believes it is acceptable to overstate insurance claims, according to a 1997 poll by the Malvern, Pennsylvania-based Insurance Research Council. Without the public's acknowledgement that insurance fraud is a crime that costs everyone, the problem will continue to grow.

7. Electronic transactions. The explosion of electronic and online transactions significantly increases the potential for fraud and the size of the losses that can occur in real-time versus day-long or week-long transactions. London-based Meridian Research projects that annual online fraud will rise to $60 billion by 2005 (including all online business transactions). A study from Stamford, Connecticut-based Gartner, Inc. supports these numbers, showing that online retailers are twelve times more likely to be defrauded than their traditional counterparts. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.