A Multi-Line Insurance Fraud Recognition System: A Government-Led Approach in Korea

Abstract

ABSTRACT This article introduces a government-led insurance fraud detection program in Korea. The Insurance Fraud Recognition System (IFRS) uses policy and claims data from multi-lines of insurance (life, automobile, and fire), employs a threestage statistical and link analysis to identify presumably fraudulent claims by claimant or by group, and generates system reports that the government regulator draws on to make decisions. The authors evaluate the system based on the fraud statistics and IFRS results for 2004, and offer recommendations for system improvement. This article examines existing studies about fraud, industry experiments using advanced technology, and government assistance to the insurance industry's fight against fraud in selected countries. It also provides a brief overview of the Korean insurance market, especially after the recent Asian economic crisis. INTRODUCTION Every valid insurance contract requires the presence of insurable interest. Without this requisite, unethical persons and entities would use the insurance mechanism to support their gambling activity rather than to protect their wealth against future losses, and insurers would face extreme difficulty in estimating their contractual liabilities to policyholders. Controlling problems of moral hazard with this requisite is a means of keeping the cost of insurance at a reasonable level. However, use of this passive approach alone does not guarantee operational and financial soundness of the insurance mechanism. Insurers need to employ active approaches to identify genuine claims and expedite services for those claims, while deterring people from filing false or inflated claims. Detecting such claims is equally important. The term "fraud" in the insurance industry is broadly used to refer to false or inflated claims. To be precise, fraud refers to an act that a person or entity, individually or jointly, willfully commits to obtain a monetary gain from an insurer by knowingly presenting false evidence of economic loss.1 The evidence can be false in its entirety, thus making the act an attempt of "hard fraud," or false in part (e.g., inflating the actual loss amount), thus making it an attempt of "soft fraud." These types of deliberate and intentional acts, when not prevented or captured, increase the cost of insurance. Ex ante elimination of fraudulency in the insurance market is feasible only in theory, as it literally means a complete control of moral hazard in the market. Instead, we tend to employ ex post approaches that can effectively deter unethical persons' attempts to gain financially by ill-using the insurance mechanism, or penalize the deceit. These approaches often require coordination of efforts by three parties. First, the insurance industry must develop a market environment where genuine claims are honored. It can do so by furthering the clarity of the terms and conditions in insurance contracts and by maintaining effective claims management programs. The government can help the industry by reforming the legal environment so that insurance fraud is a crime without exception. The academician can offer the research support for the development and enhancement of fraud deterrence or detection programs. We discuss these tripartite efforts in the section below. Academic Research Numerous researchers, some in academia and others in the industry, have, since the early 1980s, examined and proposed solutions to problems of moral hazard (and fraud in particular) in the insurance market. Notable studies include, but are not limited to, examination of how the presence of insurance affects the possibilities of fraud (Dionne, 1984), perception of fraud by the insurer and by the insured in eight countries (Clark, 1990), behavioral factors and lottery conditions under the no-fault automobile insurance system (Derrig, Weisberg, and Chen, 1994), workers' compensation insurance fraud (Butler, Durbin, and Helvacian, 1996), the theoretically fraud-inducing economic environment (Boyer, 2000), claims auditing as a means to deter and detect fraud in automobile insurance (Tennyson and Salsas-Forn, 2002), use of general damage awards by insurers to reduce fraudulent claims (Loughran, 2005), and the relationship between coinsurance and fraud frequency in the healthcare industry (Sulzle and Wambach, 2005). …