Test

This article was updated on March 6, 2024. Advances in analytics and modeling are making credit risk decisioning more efficient and precise. And while businesses may face challenges in developing and deploying new credit risk models, machine learning (ML) — a type of artificial intelligence (AI) — is paving the way for shorter design cycles and greater performance lifts. LEARN MORE: Get personalized recommendations on optimizing your decisioning strategy Limitations of traditional lending models Traditional lending models have worked well for years, and many financial institutions continue to rely on legacy models and develop new challenger models the old-fashioned way. This approach has benefits, including the ability to rely on existing internal expertise and the explainability of the models. However, there are limitations as well. Slow reaction times:  Building and deploying a traditional credit risk model can take many months. That might be okay during relatively stable economic conditions, but these models may start to underperform if there's a sudden shift in consumer behavior or a world event that impacts people's finances. Fewer data sources:  Traditional scoring models may be able to analyze some types of FCRA-regulated data (also called alternative credit data*), such as utility or rent payments, that appear in credit reports. Custom credit risk scores and models could go a step further by incorporating data from additional sources, such as internal data, even if they're designed in a traditional way. But AI-driven models can analyze vast amounts of information and uncover data points that are more highly predictive of risk. Less effective performance:  Experian has found that applying machine learning models can increase accuracy and effectiveness, allowing lenders to make better decisions. When applied to credit decisioning, lenders see a Gini uplift of 60 to 70 percent compared to a traditional credit risk model.1 Leveraging machine learning-driven models to segment your universe From initial segmentation to sending right-sized offers, detecting fraud and managing collection efforts, organizations are already using machine learning throughout the customer life cycle. In fact, 79% are prioritizing the adoption of advanced analytics with AI and ML capabilities, while 65% believe that AI and ML provide their organization with a competitive advantage.2 While machine learning approaches to modeling aren't new, advances in computer science and computing power are unlocking new possibilities.3 Machine learning models can now quickly incorporate your internal data, alternative data, credit bureau data, credit attributes and other scores to give you a more accurate view of a consumer's creditworthiness. By more precisely scoring applicants, you can shrink the population in the middle of your score range, the segment of medium-risk applicants that are difficult to evaluate. You can then lower your high-end cutoff and raise your low-end cutoff, which may allow you to more confidently swap in  good accounts (the applicants you turned down with other models that would have been good) and swap out bad accounts (those you would have approved who turned bad). Machine learning models may also be able to use additional types of data to score applicants who don't qualify for a score  from traditional models. These applicants aren't necessarily riskier — there simply hasn't been a good way to understand the risk they present. Once you can make an accurate assessment, you can increase your lending universe by including this segment of previously "unscorable" consumers, which can drive revenue growth without additional risk. At the same time, you're helping expand financial inclusion to segments of the population that may otherwise struggle to access credit. READ MORE: Is Financial Inclusion Fueling Business Growth for Lenders? Connecting the model to a decision Even a machine learning model doesn't make decisions.4 The model estimates the creditworthiness of an applicant so lenders can make better-informed decisions. AI-driven credit decisioning software can take your parameters (such cutoff points) and the model's outputs to automatically approve or deny more applicants. Models that can more accurately segment and score populations will result in fewer applications going to manual review, which can save you money and improve your customers' experiences. CASE STUDY:  Atlas Credit, a small-dollar lender, nearly doubled its loan approval rates while decreasing risk losses by up to 20 percent using a machine learning-powered model and increased automation. Concerns around explainability One of the primary concerns lenders have about machine learning models come from so-called “black box" models.5 Although these models may offer large lifts, you can't verify how they work internally. As a result, lenders can't explain why decisions are made to regulators or consumers — effectively making them unusable. While it's a valid concern, there are machine learning models that don't use a black box approach. The machine learning model doesn't build itself and it's not really “learning" on its own — that's where the black box would come in. Instead, developers can use machine learning techniques to create more efficient models that are explainable, don't have a disparate impact on protected classes and can generate reason codes that help consumers understand the outcomes. LEARN MORE: Explainability: Machine learning and artificial intelligence in credit decisioning Building and using machine learning models Organizations may lack the expertise and IT infrastructure required to develop or deploy machine learning models. But similar to how digital transformations in other parts of the business are leading companies to use outside cloud-based solutions, there are options that don't require in-house data scientists and developers. Experian's expert-guided options can help you create, test and use machine learning models and AI-driven automated decisioning; Ascend Intelligence Services™ Acquire:  Our model development service allows you to prebuild and test the performance of a new model before Experian data scientists complete the model. It's collaborative, and you can upload internal data through the web portal and make comments or suggestions. The service periodically retrains your model to increase its effectiveness. Ascend Intelligence Services™ Pulse:  Monitor, validate and challenge your existing models to ensure you're not missing out on potential improvements. The service includes a model health index and alerts, performance summary, automatic validations and stress-testing results. It can also automatically build challenger models and share the estimated lift and financial benefit of deployment. PowerCurve® Originations Essentials:  Cloud-based decision engine software that you can use to make automated decisions that are tailored to your goals and needs. A machine learning approach to credit risk and AI-driven decisioning can help improve outcomes for borrowers and increase financial inclusion while reducing your overall costs. With a trusted and experienced partner, you'll also be able to back up your decisions with customizable and regulatorily-compliant reports. Learn more about our credit decisioning solutions. Learn more When we refer to "Alternative Credit Data," this refers to the use of alternative data and its appropriate use in consumer credit lending decisions as regulated by the Fair Credit Reporting Act (FCRA). Hence, the term "Expanded FCRA Data" may also apply in this instance and both can be used interchangeably.1Experian (2024). Improving Your Credit Risk Machine Learning Model Deployment2Experian and Forrester Research (2023). Raising the AI Bar3Experian (2022). Driving Growth During Economic Uncertainty with AI/ML Strategies4Ibid5Experian (2020). Explainability ML and AI in Credit Decisioning

This article was updated on March 4, 2024. If you steal an identity to commit fraud, your success is determined by how long it takes the victim to find out. That window gets shorter as businesses get better at knowing when and how to reach an identity owner when fraud is suspected. In response, frustrated fraudsters have been developing techniques to commit fraud that does not involve a real identity, giving them a longer run-time and a bigger payday.  That's the idea behind  synthetic identity (SID) fraud — one of the fastest-growing types of fraud.  Defining synthetic identity fraud Organizations tend to have different  definitions of synthetic identity fraud, as a synthetic identity will look different to the businesses it attacks. Some may see a new account that goes bad immediately, while others might see a longer tenured account fall delinquent and default. The qualifications of the synthetic identity also change over time, as the fraudster works to increase the identity’s appearance of legitimacy. In the end, there is no person to confirm that fraud has occurred, in the very best case, identifying a synthetic identity is inferred and verified. As a result, inconsistent reporting and categorization can make tracking and fighting SID fraud more difficult.  To help create a more unified understanding and response to the issue, the Federal Reserve and 12 fraud experts worked together to develop a definition. In 2021, the  Boston Federal Reserve  published the result, “Synthetic identity fraud is the use of a combination of personally identifiable information to fabricate a person or entity to commit a dishonest act for personal or financial gain."1 To break down the definition, personally identifiable information (PII) can include:  Primary PII:  Such as a name, date of birth (DOB), Social Security number (SSN) or another government-issued identifier. When combined, these are generally unique to a person or entity. Secondary PII:  Such as an address, email, phone number or device ID. These elements can help verify a person or entity's identity.   Synthetic identities are created when fraudsters establish an identity from scratch using fake PII. Or they may combine real and fake PII (I.e., a stolen SSN with a fake name and DOB) to create a new identity. Additionally, fraudsters might steal and use someone's SSN to create an identity – children, the  elderly  and incarcerated people are popular targets because they don't commonly use credit.4 But any losses would still be tied to the SID rather than the victim. Exploring the Impact of SID fraud The most immediate and obvious impact of SID fraud is the fraud losses. Criminals may create a synthetic identity and spend months  building up its credit profile, opening accounts and increasing credit limits. The identities and behaviors are constructed to look like legitimate borrowers, with some having a record of on-time payments. But once the fraudster decides to monetize the identity, they can apply for loans and max out credit cards before ‘busting out’ and disappearing with the money.  Aite-Novaric Group estimates that SID fraud losses totaled $1.8 billion in 2020 and will increase to $2.94 billion in 2024.2 However, organizations that do not identify SIDs may classify a default as a credit loss rather than a fraud loss.  By some estimates, synthetic identity fraud could account for up to 20 percent of loan and credit card charge-offs, meaning the annual charge-off losses in the U.S. could be closer to $11 billion.3 Additionally, organizations lose time and resources on collection efforts if they do not identify the SID fraud.  Those estimates are only for unsecured U.S. credit products. But fraudsters use synthetic identities to take out secured loans, including auto loans.   As part of schemes used to steal relief funds during the pandemic, criminals used synthetic identities to open demand deposit accounts to receive funds. These accounts can be used to launder money from other sources and commit peer-to-peer payment fraud. Deposit account holders are also a primary source of cross-marketing for some financial institutions. Criminals can take advantage of vulnerable onboarding processes for deposit accounts where there’s low risk to the institution and receive offers for lending products. Building a successful SID prevention strategy Having an effective SID prevention strategy is more crucial than ever for organizations. Aside from fraud losses, consumers listed identity theft as their top concern when conducting activities online. And while 92% of businesses have an identity verification strategy in place, 63% of consumers are "somewhat confident" or "not very confident" in businesses' ability to accurately identify them online. Read: Experian's 2023 Identity and Fraud Report Many traditional fraud models and identity verification methods are not designed to detect fake people. And even a step up to a phone call for verification isn't enough when the fraudster will be the one answering the phone. Criminals also quickly respond when organizations update their fraud detection methods by looking for less-protected targets. Fraudsters have even signed their SIDs up for social media accounts and apps with low verification hurdles to help their SIDs pass identity checks.5  Understand synthetic identity risks across the lifecycle  Synthetic Identities are dynamic. When lending criteria is tightened to synthetics from opening new accounts, they simply come back when they can qualify. If waiting brings a higher credit line, they’ll wait. It’s important to recognize that synthetic identity isn’t a new account or a portfolio management problem – it’s both.    Use analytics that are tailored to synthetic identity  Many of our customers in the financial services space have been trying to solve synthetic identity fraud with credit data. There’s a false sense of security when criteria is tightened and losses go down—but the losses that are being impacted tend to not be related to credit. A better approach to synthetic ID fraud leverages a larger pool of data to assess behaviors and data linkages that are not contained in traditional credit data.  You can then escalate suspicious accounts to require additional reviews, such as screening through the Social Security Administration's Electronic Consent Based SSN Verification (eCBSV) system or more stringent document verification.  Find a trusted partner  Experian's interconnected data and analytics platforms offer lenders turnkey identity and synthetic identity fraud solutions. In addition, lenders can take advantage of the risk management system and continuous monitoring to look for signs of SIDs and fraudulent activity, which is important for flagging accounts after opening. These tools can also help lenders identify and prevent other common forms of fraud, including account takeovers, e-commerce fraud, child identity theft fraud and elderly fraud. Learn more about our synthetic identity fraud solutions. Learn more 1Federal Reserve Bank (2021). Defining Synthetic Identity Fraud 2Aite Novarica (2022). Synthetic Identity Fraud: Solution Providers Shining Light into the Darkness 3Experian (2022). Preventing synthetic identity fraud 4The Federal Reserve (2022). Synthetic Identity Fraud: What Is it and Why You Should Care? 5Experian (2022). Preventing synthetic identity fraud 

This series will dive into our monthly State of the Economy report, providing a snapshot of the top monthly economic and credit data for those in financial services to proactively shape their business strategies. In February, economic growth and job creation outperformed economists’ expectations, likely giving confirmation to the Federal Reserve that it remains too early to begin cutting rates. Data highlights from this month’s report include: U.S. real GDP rose 3.3% in Q4 2023, driven by consumer spending and bringing the average annual 2023 growth to 2.5%, the same as the five-year average growth prior to the pandemic. The labor market maintained its strength, with 353,000 jobs added this month and unemployment holding at 3.7% for the third month in a row. Consumer sentiment rose 13% in January, following a 14% increase in December, as consumers are feeling some relief from cooling inflation. Check out our report for a deep dive into the rest of February’s data, including inflation, the latest Federal Reserve announcement, the housing market, and credit card balances. To have a holistic view of our current environment, we must understand our economic past, present, and future. Check out our annual chartbook for a comprehensive view of the past year and register for our upcoming Macroeconomic Forecasting webinar for a look at the year ahead. Download report Register for webinar For more economic trends and market insights, visit Experian Edge.