NATIONAL LAW UNIVERSITY, JODHPUR

School of Insurance Studies

LIABILITY AND AVIATION INSURANCE

ASSIGNMENT ON:
STUDY AND DETECTION OF CLAIMS FRAUD IN
INSURANCE INDUSTRY

Submitted To: Submitted By:

Dr. R.K. Purohit Khusboo Balani
School of Insurance Studies MBA Insurance- IV
Roll No. 1060

2018-20
Insurance fraud is a serious and growing problem, and there is widespread recognition that
traditional approaches to tackling fraud are inadequate. Studies of insurance fraud have
typically focused upon identifying characteristics of fraudulent claims and claimants, and
this focus is apparent in the current wave of forensic and data-mining technologies for fraud
detection. An alternative approach is to understand and then optimize existing practices in the
detection of fraud. In a world where insurance is both a form of investment and security,
many try to take advantage of the policies and systems that revolve around it. One of the
ways that perpetrators try to illegally get money from insurance companies is through
insurance claims fraud. Insurance fraud can both be done personally or through online
claims. This is why cyber security is crucial for many businesses. Globally, insurance fraud is
a major concern for Insurers which continues to increase year by year. Claims fraud is the
most common buzz around the Insurance industry with auto and workers compensation
business segments being the major contributors. Frauds are typically an individual or a group
led effort of fraudsters with an intention of inflating claims and finally making profit out of a
loss. Insurers spend huge effort and manpower in detecting fraud which as a net result not
only drain the dollar amount from the insurer’s kitty but also adversely affect the good risks
that they underwrite. It is also a social risk as it promotes financial crimes and penalizes the
society.

What is Insurance Claims Fraud?

By definition, insurance fraud is when a claimant makes up information or exaggerates
information in order to get the money guaranteed by their insurance policy. As a result, these
false insurance claims can lead to increased rates, taxes, and inflated policy prices for your
clients. As there are various types of insurances, there are also various types of devious
techniques that scammers do to get money out of these policies. Business owners must be
vigilant on how they review insurance claims in order to detect which ones are legitimate or
not.

1
Common Types of Insurance Claims Fraud
Accident Insurance Fraud
Some scammers will try to stage accidents in order to claim a large sum of money from
insurance companies. They will claim fake injuries, fake accidents, or stage accidents
towards themselves or family members in order to get the claim. Additionally, these types of
fraudulent activities can also involve a crime of some type, although it appears on the surface
as an accident. At this type of accident frauds, you need to have police authorities involved in
the investigation.
Contractor Insurance Fraud
This is the type of fraud that is not usually done by the policyholder but rather by the
contractor of a construction business. Some contractors will declare a price of goods and
services, which can inflate the price of the home insurance. As a result, this fraudulent
activity may cost the policyholder more in his insurance while the contractor takes home the
money for their low-quality goods or services. When dealing with this type of fraud, it is
important to have regular inspections and investigations of the contractors involved in
building the home.
Break-in Insurance Fraud
Another type of fraud that involves home insurance is staged break-ins. At this type of fraud,
the homeowner tries to claim that there was a burglar that ruined their property or stole some
valuable items. Others would try to commit arson and pretend it was some unknown criminal
who set the house on fire. When a homeowner is caught committing these types of fraud, it
can lead to criminal charges. Insurance companies must also be vigilant with these types of
claims.
Disaster Insurance Fraud
Many insurance companies offer policies that involve house repairs to the provision of
monetary resources when a disaster strikes. Although policyholders can be honest, some
operators would try to convince or connive with them in order to get a higher insurance
claim. They would make false reports about the extent of the disaster, thus scamming the
insurance company with the value that needs to be provided.

Following are the key aspects around claim fraud, due to which
imperatives of the insurance ecosystem are impacted in different degrees of
severity:-

2
 Underwriting: Claims Fraud impacts underwriting guidelines and policies and
deteriorates the insurance risk pool.
 Social Costs: Due to claim fraud, the prices of insurance go up as a whole.
 Unfair with rightfully deserving: Due to claim fraud sometimes even the rightfully
deserving claimants are either denied claim or have to provide additional proofs on a claim.
 Undetected fraud encourages more fraud: When to a successful fraud propensity of
individuals to further indulge in fraud increases, thus encouraging more fraud.
 Loss in Reputation: Repetition of fraudulent claims for an insurance company causes
loss in market reputation thus causing a decline in competitiveness.
 Customer Relationship: Fraudulent claims adversely impacts insurer’s relationship
with its existing customers and with prospects.
 Regulatory Compliance: Repeated frauds or even a few major frauds might cause
serious legal issues for an insurer with the regulators.
 Loss of faith: Due to fraudulent claims people’s trust in insurance declines, which is
detrimental to the growth of insurance industry.

Detection of claim fraud in insurance industry

The detection of insurance fraud generally occurs in two steps. The first step is to identify
suspicious claims that have a higher possibility of being fraudulent. This can be done
by computerized statistical analysis or by insurance agents. Additionally, the public can
provide tips to insurance companies, law enforcement and other organizations regarding
suspected, observed, or admitted insurance fraud perpetrated by other individuals. Regardless
of the source, the next step is to refer these claims to investigators for further analysis.
Due to the sheer number of claims submitted each day, it would be far too expensive for
insurance companies to have employees check each claim for symptoms of fraud. Instead,
many companies use computers and statistical analysis to identify suspicious claims for
further investigation. There are two main types of statistical analysis tools used: supervised
and unsupervised. In both cases, suspicious claims are identified by comparing data about the
claim to expected values. The main difference between the two methods is how the expected
values are derived.
In a supervised method, expected values are obtained by analyzing records of both fraudulent
and non-fraudulent claims. According to Richard J. Bolton and David B. Hand, both of
Imperial College in London, this method has some drawbacks as it requires absolute certainty
that those claims analyzed are actually either fraudulent or non-fraudulent, and because it can
only be used to detect types of fraud that have been committed and identified before.
Unsupervised methods of statistical detection, on the other hand, involve detecting claims
that are abnormal. Both claims adjusters and computers can also be trained to identify "red
flags", or symptoms that in the past have often been associated with fraudulent
claims. Statistical detection does not prove that claims are fraudulent; it merely identifies
suspicious claims that must be investigated further.

3
 Fraudulent claims can be one of two types:

 they can be otherwise legitimate claims that are exaggerated or "built up", or
 they can be false claims in which the damages claimed never actually occurred.
Once a built up claim is identified, insurance companies usually try to negotiate the claim
down to the appropriate amount. Suspicious claims can also be submitted to "special
investigative units", or SIUs, for further investigation. These units generally consist of
experienced claims adjusters with special training in investigating fraudulent claims. These
investigators look for certain symptoms associated with fraudulent claims, or otherwise look
for evidence of falsification of some kind. This evidence can then be used to deny payment of
the claims or to prosecute fraudsters if the violation is serious enough.
When an insurance company's fraud department investigates a fraud claim, they frequently
proceed in two stages: pre-contact and post-contact. In the pre-contact stage they analyze all
available evidence before they contact the suspect. They may reviewing submitted
paperwork, reach out to third parties, and gather evidence from available sources. Then, in
the "post-contact" stage, they interview the suspect to gather more information and, ideally,
obtain an incriminating statement. Insurance fraud investigators are trained to question the
suspect in a way that precludes the suspect raising a valid defense at a later time. For
example, questions about access to claim forms preclude the defense that another individual
filled out the fraudulent documents. Full disclosure may add credibility to a suspect's account
of events, but omissions from disclosure or false statements may detract from the suspect's
credibility in later interviews or proceedings.

A) Detection of insurance fraud with the help of INSURANCE AGENTS

in following ways:-
1. Analyze Claims History
Insurers also try to discern any patterns in your past claims regarding their frequency and
type. You may not realize it, but insurance companies keep in-depth records on claims and
do all sorts of analyses to interpret the data they contain -- everything from figuring out who
is most likely to file a claim to when and where. If your claim doesn't match the typical
pattern, they'll notice.
2. Check List of Suspicious Loss Indicators

Few of the suspicious loss indicators insurance agents look out for:-

 A claimant who's totally calm and unflustered after submitting a large claim
 A claimant who submits handwritten receipts for repairs on a covered item
 A claimant who adds or increases homeowners or auto insurance coverage shortly
before submitting a claim
 A fire-damage claim for a home or auto where the fire started immediately after a
family argument, or shortly after family members left the home/car
 A medical claim submitted by a seasonal employee whose job is ending

4
Of course, some of these scenarios can be present in legit claims. But. Insurers know they're
not positive indicators of fraud just possible ones and definitely signs they should further
investigate certain claims.

3. Use Private Investigators

Private investigators use tactics to uncover fraud, such as researching claimants' backgrounds
through perusing criminal records, interviewing claimants and any witnesses, inspecting
pertinent sites, etc. Investigators may also consult with the insurer's legal counsel and serve
as expert witnesses in court. While some private investigators are hired by insurance
companies on a freelance basis, insurers also hire in-house PIs, often selecting those with
backgrounds in law enforcement and private investigation .
4. Look for Evidence of Personal Injury Mills
One of the more popular insurance fraud scams involves vehicle crashes that result in both
legitimate and fake/exaggerated injuries. Typically, those involved in such practices -- certain
medical providers or lawyers -- will perform the same scam over and over. If insurers notice
a particular provider submitting numerous claims over time for accident victims that
coincidentally receive a similar treatment regimen, that's red flag.
5. Use Sophisticated Computer Systems to Detect Suspicious Billing
Fraud often occurs through billing, and quite often for medical claims. Physicians or clinics
may bill insurance companies for services never rendered, for example, or for procedures or
services that weren't medically necessary. Or they may jack up the cost of certain services,
charge more than once for the same service or "unbundle" claims -- billing, say, for three
separate surgeries on a patient whose three toes were operated on at the same time. Complex
computer systems have been developed to tease out suspicious bills and billing patterns from
physicians and medical.
6. Hand Case to Special Investigation Unit
Many insurers have Special Investigation Units, or SIUs. Employees who work in SIUs
generally have backgrounds as detectives, police officers, medical personnel and the like.
They're capable of performing an amazing array of tests and checks to bust anyone trying to
commit fraud.
7. Evaluate Prospective Employees' Credit Histories
Insurance fraud isn't limited to external sources. There's a certain amount of it that originates
with insurance companies' own staffs. Claims adjustors cut a lot of checks, and unethical
folks may try to skim a few bucks off the top. Agents can commit fraud by "stealing"
customers' car insurance or life insurance premiums. The agents take in customers' money,
then pocket it without ever actually purchasing the policies. Savvy insurance companies try to
prevent such fraud by running credit checks on all prospective employees. Applications from
those with bad credit or financial issues are flagged as those more likely to commit fraud.

5
8. Check on Claimants Through Social Media
Insurers are now using social media to check up on suspicious claims. Perhaps the claimant
who said his car suffered hail damage will be bragging about his deception on Facebook or
Twitter, or will upload a video on YouTube showing how to create fake hail dents in your
car's hood. Social media's biggest application in claims fraud, though, is in disability cases. A
quick look at a claimant's Facebook photos and postings often makes it clear whether the
person is truly disabled.
9. Perform Cross-Checks
One of the easiest ways for insurers to catch crooks is via a basic cross-check. All this
involves is looking for simple patterns in the checks they're sending out to pay claims. If the
same person is receiving numerous checks, that's a warning sign. So is the payment of several
big claims to the same address, even if the name on the check is different.Cross-checks aren't
limited to an insurer's database, either. Thousands of insurance companies, self-insured
entities and third-party administrators report all of their claims to ISO ClaimSearch, an anti-
fraud information system.

B) Detection of insurance fraud with the help of COMPUTERIZED

STATISTICAL ANALYSIS  in following ways:-
1: Gearing Up for Data Management
Data is the most valuable commodity for detecting and preventing insurance fraud. Insurers
have multiple legacy transactional systems, often for each line of business. Insurers are now
using advanced analytical techniques such as probabilistic matching to determine the
actual statistical likelihood that two entities are the same. Companies should not
underestimate the amount of work required for data management. An enterprise data
architecture can standardize and rationalize data across the enterprise, address data
quality and accessibility, easily incorporate external data, and include mechanisms to
capture and use both structured and unstructured data.
2: Visualizing Data
Data visualization enables insurers to identify patterns and relationships in data that
weren’t initially evident. Interactive, self-service business intelligence and reporting
capabilities are combined with out-of-the-box advanced analytics so everyone can
discover insights from any size and type of data, including text. Data analysts and special
investigation units (SIUs) can develop more effective strategies through speeding up time
to detection, identifying new insights and predicting emerging fraud trends quicker.
3: Harnessing Business Rules
Rules-based systems test each transaction against a predefined set of algorithms or business
rules to detect known types of fraud based on specific patterns of activity. These systems
flag any claims that look suspicious due to their aggregate scores or relation to threshold
values. The advantage of the flag approach is its simplicity. After initially configuring the
business rules, it is easy to match activities to accounts with very little investment or training.
Unfortunately, business rules often generate high false-positive rates and undetected fraud
because fraudsters can easily learn and manipulate such rules.

6
4 : Searching Databases
Insurance companies have been relying on internal data to identify and fight claims for
decades. But often this is not enough to recognize suspicious activities, and capitalizing on
existing external data has become crucial. Adding sources such as industry consortiums that
aggregate and share historical claims data to validate new claims, fraud watch lists, public
records, medical billing data, underwriting information and auto estimates can have
tremendous impact on fraud modelling.
5: Detecting Anomalies
With anomaly detection, you define baselines for key performance indicators (KPIs)
associated with tasks or events, then set thresholds. When a threshold for a particular
measure is exceeded, then the event is reported. Outliers or anomalies could indicate a new
or previously unknown pattern of fraud.
This analytical technique is straightforward, easy to implement, and – besides detecting unknown
fraud patterns – can be useful for evaluating individual performance and identifying employee
training opportunities. Once in place, the system functions automatically. Adjuster and SIU
activities are monitored, and you can quickly identify and correct problems.
6: Delving Deeper With Predictive Modelling
A predictive modelling process is where data scientists, statisticians and quantitative analysts
use data mining tools to build predictive models that produce fraud propensity scores. As
data is entered and updated, claims are automatically scored for their likelihood to be
fraudulent and made available for review. Predictive modelling tends to be more accurate
than other fraud detection methods. Information can be collected and cross-referenced from a
variety of data sources. This diversity of resources provides a better balance of data than the
more labor-intensive business rules flagging system.
7: Realizing the Value of Text Analytics
Within many insurance organizations, some of the most useful information is buried within
unstructured textual data. The claims process collects and generates large volumes of text-
based information, such as adjuster notes, emails, customer service calls and claimant
interviews. In fact, unstructured data can represent up to 80 percent of claims data.
Text mining software accesses the unstructured text, parses it to distill meaningful data, and
analyzes the newly created data to gain a deeper understanding of the claim. For example,
you might use text mining to look for scripted comments in auto-accident claims. It would
be a little suspicious if multiple claimants, allegedly unrelated, all say exactly the same thing.
Text mining can be very helpful in revealing these types of discrepancies or conditions.
8: Identifying Organized Fraud Through Social Network Link Analysis
Network link analysis has proven effective in identifying organized fraud activities by
modelling relationships between entities in claims and new business. The key entity is
individual or claimant, but other types of entities may be locations, service providers,
telephone numbers and vehicle identification numbers – to name just a few. Large
volumes of seemingly unrelated claims can be checked, and then patterns and problems
identified. Insurers have successfully used network link analysis to identify the presence

7
of organized fraud rings and take appropriate action. Furthermore, insurers can use these
linking and network scoring techniques to not only avoid paying fraudulent claims at first
notification of loss, but also check new policies for connections to historical fraud to
avoid proliferation of fraud.
9: Managing and Triaging Alerts
The management of suspicious activities and fraud cases is critical to the business, but also
very time-sensitive. Rules often regulate how quickly an insurance company must respond
to a claim, resulting in overwhelmed SIU experts and potential fraudulent activities going
uninvestigated. Alert management assembles and prioritizes alerts and routes them to the
appropriate team member. This provides investigators with a more complete perspective on
the risk, enabling them to conduct more efficient, effective investigations, leading to reduced
costs and most importantly, better fraud prevention.
10 : Knowing Your Deployment Options
Network link analysis should be fully automated, with the system continuously updating the
interrelated networks with new claims and policies and rescoring for fraud. If a network
score indicates fraud, then this can be used to flag the new claim as it is notified and the
system matches it to the network. Investigators can search across the full customer base of
claims and policies in seconds and turn up visual indications of connections and overlaps
among them. Insurers have successfully used network link analysis to identify the presence
of organized fraud rings and take appropriate action. Furthermore, insurers can use these
linking and network scoring techniques to not only avoid paying fraudulent claims at first
notification of loss, but also check new policies for connections to historical fraud to avoid
proliferation of fraud.

Cases of claims fraud in insurance

Exaggerating damages or loss
Case 1 – Overcharge for damage repair
A report published by the California Bureau of Automotive Repair in 2002 indicated that of
over 500 vehicles inspected after repairs, more than 40% of the bills included charges for
work never done or for parts not used. The average overcharge was $586, (one-sixth of the
average auto insurance claim after an accident).

Staging the occurrence of incidents

Case 2 – Staging a car accident after illegal racing
A new car under comprehensive motor cover is used in illegal car racing, which depreciates
its value rapidly. The policyholder stages a car accident in the presence of independent
witnesses. He would then claim compensation from the insurer for damage to his car.

Reporting and claiming of fictitious damage

Case 3 – False mobile phone thefts

8
In Britain the police force receives 160 false reports of mobile phone thefts a month, which
costs it £1 million a year to investigate. The National Mobile Phone Crime Unit estimates
that between 15-20 per cent of mobile phone theft reports in the UK are false. Police
suspect that false claims are sometimes encouraged by unscrupulous mobile phone shop
staff looking for extra commission. Sometimes someone who has lost their phone will
falsely report it as stolen in order to claim on their insurance. People think they're doing
nothing wrong in lying to police and insurers.

Fraud by a third party involved in the settlement of the claim

Case 4 – Independent adjuster arrested in shakedown scheme

Mr. B., an independent adjuster, was hired by an insurer to conduct an inventory at a retail
department store, after the store had been burglarized. The owner of the store, who
cooperated with the investigation, had reported a loss of $33,599 to his insurer. The
investigators electronically monitored conversations between the owner and Mr. B, wherein
Mr. B. stated that he had figured the loss to be much lower than reported, but offered to
“inflate’ his inventory in return for 7%. B. agreed to a cash payment of $2,000. When Mr.
B. was overheard accepting the payment from the owner, he was placed under arrest.

The Need Of The Hour

Insurance fraud is multifaceted. There is no single typical profile of the fraudulent claimant.
Fraud is committed by polished professional criminals and ordinary citizens. A single act of
fraud may involve millions of dollars or less than one hundred dollars. Increased law
enforcement may be the way to deter or catch professional defrauders, but small-time fraud
must be stopped by efforts within the insurance industry. Lack of a proper penal code to
punish fraudsters hinders the eradication of frauds from the insurance industry. Stringent laws
and regulatory policies can play a significant role in curbing frauds.
Emerging technologies such as artificial intelligence (AI) and block chain has the potential to
smartly bring down insurance frauds. Use of machine learning for fraud detection means that
AI learns to improve their results over time, getting the ability to notice the signs of fraud
more efficiently as they encounter any new instances. A holistic approach towards fraud
control supported by technological and regulatory measures will support India’s vision of
securing a stronger penetration for insurance and protecting people.