Insurance and fraud: what is the cost?

Fraud consists of a deliberate action designed to unlawfully gain from an insurance policy. It is usually committed at the time of subscription or by means of a false declaration as to the causes, circumstances or consequences of the claim. 

Insurance fraud figures

In 2015, insurance fraud cases represented a total estimated cost of 2.5 billion euros in damages, i.e. close on 5% of the premiums collected each year for car insurance and property, fire, accidents and miscellaneous risks (“IARD”).  

According to the 2015 report of the French agency for the fight against insurance fraud (ALFA), the analysis of which covered 78% of the French IARD market, general insurance companies identified some 46,255 cases of fraud, a figure 8% up over 2014 and more than 4 times higher than that of 2003. The action of the anti-fraud departments of insurers enabled savings of 265 million euros in payments not made to fraudulent policyholders, of which one half concerned car insurance. Projected over 100% of the French market, ALFA estimates the total annual number of frauds detected to be close on 60,000 for just over 339 million euros to be recovered.

This increase may be explained in part by the strategies implemented by insurers to combat fraud enabling a more efficient detection of the relevant cases to identify it.

ALFA has established a link of cause and effect between the explosion in minor insurance fraud and the economic crisis, and the massive influx of fraud through third party insurance service providers (opticians, garages, breakdown services, etc.).

Finally, if the number of cases of fraud has shown to be constantly on the rise, it is also because insurers are now implementing increasingly effective strategies to combat fraud.

Number of cases of insurance fraud detected by insurers:

2003: 11,937

2011: 35,042

2015: 46,255

Breakdown of fraudulent dossiers according to the type of claim in 2015:

Theft: 16 %

Multi-risk civil liability: 7 %

Fire: 5 %

Other risks: 72 % 

Fraud detection: a real challenge for insurers

Fraud is an operational risk within the meaning of the Solvency 2 Directive which came into force in France on 1st January 2016. It must therefore be part of the inventory of insurers with adequate provisioning and measures designed to cope with it.  

Over the last 10 years, in their fight against fraud insurers have reinforced their internal control resources and have acquired a wide range of tools designed to ensure better detection of cases of fraud, whether on an individual level or as organized crime, the most costly form.  Fraud generally comes in the form of false claims, the existence of which must be formally proved by the insurer as the policyholder is always presumed to be acting in good faith. For such insurers, the aim is to boost reports of suspicious situations, to reinforce controls and to recover unwarranted pay-outs while at the same time reducing investigation and court costs.  

The means deployed by insurers to detect fraud aim essentially at considerably heightening the awareness of the problem among all the internal players (subscription services, claims managers, computer experts, jurists, etc.) and external players, such as the networks of assessors and repairers.  For example, in the case of handling a car theft claim, fraud detection is part and parcel of the assessment operations.  

Such human resources are backed by computerised data analysis tools (tools for the global management of policies, tools for the validation of subscriptions, tools for the parameterisation of fraud scenarios, tools for interconnection with data in one and the same portfolio) in order to detect fraudulent undertakings.   

In the future, fraud detection in insurance companies may well make a major step forward thanks to the development of big data and to predictive analysis, based on the one hand on the computerised storage and processing of data from insurers and on the other hand on information drawn from the public digital channels (such as social media), with due observance of the statutory provisions governing data processing. The objective is to create automatic detection scenarios especially for fraud as organised crime.

The sanctions for insurance fraud

In the event of fraud at the time of subscription, the Insurance Code provides for the voidance of the insurance policy and for the reimbursement by the policyholder of the compensation paid by the insurer for all claims prior to the discovery of the false declaration. 

In the event of a false declaration at the time of a claim, the sanction can only be contractual, for lack of any specific provision in the Insurance Code. The majority of policies therefore provide for the forfeiture of the policyholder’s entitlement to the guarantee in the event of any deliberately false declaration, irrespective of the causes, the circumstances or the consequences of the claim.

Finally, there is no specific criminal sanction in punishment of insurance fraud. The provisions of the Criminal Code governing embezzlement or falsification of accounts are applicable. A sector on which startup are already acting.    

Sources: ALFA (Agency for the fight against insurance fraud) – L’Argus de l’Assurance