lending fraud

A post-pandemic drive for growth in the banking sector may be fueling the growing problem of loan fraud. But can advanced analytics solve the problem?


Post-pandemic, there have been three steadily increasing types of lending fraud that are presenting significant challenges for financial institutions worldwide: application fraud, longer-term structured lending fraud such as mortgage fraud, and the highly organized “bust out fraud.”


Fraud perpetrators have become increasingly agile, taking advantage of banks who have faced challenging operational circumstances during the pandemic. Despite most institutions having anti-fraud divisions, the ability to bring together the right expertise and data presents a real challenge due to the sheer volume and scale of portfolios and diversity of products.  


However, by applying advanced analytics and using data-driven approaches, the challenges around lending fraud can be addressed – and solved – effectively.


But first, let us take a closer look at the three types of lending fraud in more detail:


Loan Fraud

In application fraud, criminals apply for new bank accounts with attractive pre-approved benefits including overdrafts, loan facilities and/or credit cards, which allow funds to be withdrawn quickly. But the fraudsters opening these accounts have no intention of repayment. In most cases, the fraudsters apply using “synthetic identities” – this involves using a mix of legitimate and fraudulent information; some representing stolen personal data from real consumers, and some fabricated by the fraudsters.


Several factors have contributed to the growth of application fraud:

  • The growth of digital channels for bank account applications provides a risk-free method of applying for new accounts without fear of being caught. If applications are denied, the fraudsters learn from each attempt and try again until they succeed.
  • The pressure on banks to create fast and frictionless new customer application and approval processes is extremely competitive and can be measured in the number of clicks required and time taken.
  • The wide availability of personal identity information stolen in large-scale data breaches on the black market.
  • Economic pressures created by the Coronavirus pandemic, among other factors.

Mortgage Fraud

Around the millennium, mortgage fraud cases were considered a minimal threat to the banking sector. Those that were reported typically involved a small number of mortgaged properties with equally small potential losses. They would include networks using the simple typologies of boosted incomes and the provision of false information to obtain lending.  


A decade later, mortgage fraud reported by the banking sector had evolved to include highly organized mortgage frauds, some featuring portfolios of hundreds of properties and resulting in huge losses. 


However, the methods used to deceive lenders remained similar; it was being achieved using boosted incomes and false documents. The organized crime networks tended to involve corrupt business professionals who want to take advantage of generous low deposit mortgage schemes. The result? A number of highly leveraged mortgages obtained through the skill of deception. 


Now in 2021, the fraud challenge lingers. The drive for post-pandemic growth with new or existing customers has resulted in criminals seeking long term structured loans such as mortgages, which represent the prospect of a steady stream of profits for banks over decades.  


Today, mortgage fraud can still involve applicants fraudulently overstating their incomes and providing false documents. In some cases, they will do so themselves, acting alone.  However, the potential for organized crime creating networks of mortgage fraud involving large numbers of properties remains a threat. It is often the case that when these loans go delinquent, fraud is not detected as banks often categorize these losses as the result of credit risk.


Bust Out Fraud

“Bust out fraud” is a growing fraud type that is hemorrhaging losses considered to be far more significant than application and mortgage fraud. As this is a highly organized and structured type of fraud with a high degree of planning designed to deceive lenders, cases tend to run into the many millions.


In bust-out fraud cases, criminals set up or purchase existing shell companies. They then mimic the activities of a legitimate business by creating financial transactions with fake invoices and payments, creating the illusion of payroll activity, and establishing accounts with legitimate businesses. After establishing multiple accounts and credit lines and building trust, the criminals maximize their withdrawals and “bust out” – then disappear. The victims in these cases are typically other businesses and financial institutions.


As one example of the size, scale and sophistication of bust-out fraud cases, consider the successful investigation announced in February by Europol. It reported that it had arrested 105 individuals in multiple countries, who used a bust-out fraud scheme to steal an estimated €12 million ($14 million U.S.) from banks in the U.S.


Preventing Loan Fraud Can Be a Challenge 


One of the challenges in preventing lending fraud is that banks often have difficulty in determining whether the losses are credit losses or fraud, so attribution and measurement are a problem.


One leading industry analyst firm recently estimated that as much as 40% of bank write-offs were erroneously classified as credit losses when in fact, they were synthetic identity fraud. Aite Group, a financial services industry research firm, said this type of fraud “is rising to epic proportions but [it is one] that many have yet to acknowledge.”


One of the most effective methods to combat all these types of fraud is to apply advanced analytics and use data-driven approaches.


Combatting Lending Fraud with Advanced Analytics 


In the case of lending fraud, financial institutions can use analytics technology to score each new online application for fraud risk, not just credit risk. A wide range of internal and publicly available data can be used, ranging from social media data to physical addresses to email accounts, IP addresses, phone numbers, and more. Network graphing and link analysis can indicate links to other entities that indicate fraud risk, while also revealing patterns of behavior and anomalies that indicate fraud risk.


Banks and government agencies are also recognizing the value of sharing data to help indicate fraud risk. In the U.S., one notable new development in this area was taken by the Social Security Administration in 2020. With its eCBSV system, the agency began allowing banks to have access to Social Security numbers and limited personal identity information in order to score customer applications for fraud risk.


Challenges remain with this approach, such as the fact that telecommunications companies are not currently allowed access, meaning fraudsters often exploit this by establishing telecom accounts, gaining credit on those accounts, and using them to build the appearance of legitimacy for the synthetic identities they create.


When managed well, analytics can significantly reduce fraud losses from both application fraud, mortgage fraud, and business lending fraud such as bust out fraud, as well as other types of fraud. And it can do this without hampering a banks’ ability to process new customer applications in a timely manner and to deliver a successful customer experience for consumers.

Find out how to use Fraud detection software in Insurance firms

See how adding context to your fraud detection approach helps you reduce fraud losses, improve customer satisfaction and decrease the risk of reputational damage.

You may be interested in…

Better decisions start here

See how our Contextual Decision Platform transforms every operational decision you make.

Related Solutions

We use cookies to provide visitors with the best possible experience on our website. These include analytics, functionality, and targeting cookies, which may also be used in our marketing efforts. Learn about how cookies are used here.