Optimizing Credit Risk With Underwriting Technology

Woman showing underwriting data over a desk.

Trust and statistics

These two words are the foundation of underwriting. 

The concept of extending credit to an individual or business has always been based on sufficient trust that the debt could be repaid. Often, if there wasn’t trust between the parties, a guarantee or collateral would be required. 

Eventually, lenders realized that there were common characteristics of trustworthy borrowers and those who regularly defaulted. 

A merchant sailing a known sea route to a well-established colony would definitely be at lower risk than a voyage to unexplored territory. If successful, the return on investment would be higher, but the likelihood of catastrophic loss would also be higher. 

And that’s where statistics comes into play.  

The ability to accurately discern the types and magnitude of risk is a key factor in running a successful lending business. We call it underwriting.

And modern underwriting is a sophisticated process that relies heavily on statistical models. 

That’s why new data sources and analytics are poised to transform underwriting technology and unlock faster, more accurate credit risk workflows.

In this article, we’ll examine how advanced technology is changing underwriting and how your organization can use it to grow and increase profitability.   

The Cornerstone of the Lending Process

What happens when underwriting relies on incomplete data or outdated models? Businesses approve risky customers and reject qualified applicants without realizing it. 

Since the passing of the Equal Credit Opportunity Act (ECOA) and the subsequent consolidation of credit reporting agencies, underwriting methods have changed dramatically. Regulators require underwriters to remove as much subjective bias as possible and focus on rigorous qualitative risk factors. 

However, society has changed a lot in the last 50 years and will continue to change. Underwriting standards need to change with it. 

Robust, dynamic underwriting allows lenders to help drive the economy by supporting small businesses and responsible individuals who might have been overlooked by conventional standards. Today, that includes people such as gig workers, freelancers, and immigrant workers. 

Although nobody likes it when their credit application gets rejected, a fast, fair, and transparent underwriting process can help build consumer trust in your organization. Your lending operation needs a strong foundation if you’re going to stay competitive.

Advanced underwriting is the cornerstone of that foundation. 

Underwriting: A Historical Perspective

Underwriting is a deeply pragmatic business discipline. If a lender or insurance company does a poor job of it over a long enough time, they go bankrupt. In the early days, underwriting was subjective because of limited access to data. 

The character and reputation of the borrower were a major factor in the decision process. Just think of how we use the term “credibility” to refer to a person who is trustworthy.

As underwriters gained access to more data and developed bespoke processes for assessing creditworthiness, the entire economy has benefited. Lending supports growth, and insurance offers a bulwark against catastrophe. 

Technology has always been at the heart of this process because it allows underwriters to manage larger databases and perform complex analyses with greater ease.  

The Birth of Risk Assessment

There is evidence that insurance on merchant activities has existed since ancient Babylon and continued through the Roman Empire.

In fact, modern underwriting (i.e., risk assessment for the purpose of providing financial surety) owes its name and origin to a late 17th-century coffee shop in central London established by Edward Lloyd.

The foundation of Lloyd’s success was built on the shop’s reputation for reliable information about the shipping industry, including a “list” that it published using a paid subscription model.

By the 19th century the practice of underwriting grew more sophisticated as it expanded into new industries.

Statistical methods have advanced as well, thanks to professional actuaries who specialize in assessing risk and uncertainty so that banks and insurance companies can accurately price loans and policies.

Standardized Credit Evaluation

Invented by Lewis Tappan in 1841, the practice of keeping a credit file on individuals morphed into what we now know as credit history and credit score. In fact, Dun and Bradstreet, one of the largest credit reporting agencies in the world, can trace its origin directly to Lewis Tappan. 

Up until the 1970s, credit reporting was outrageously fragmented, with individual businesses maintaining confidential paper reports on their customers and more than 2000 credit bureaus.

In 1971, the Fair Credit Reporting Act (FCRA) was passed to help standardize consumer credit reporting. By the turn of the century there were only three credit bureaus—the ones we know today: Experian, Equifax, and TransUnion.

These three bureaus use a version of the credit-scoring methodology invented in 1956 by Bill Fair and Earl Isaac. Their invention gave birth to the Fair, Isaac, and Company, which became FICO. Fair and Isaac’s approach to scoring paved the way for underwriters to eliminate bias and subjectivity in assessing creditworthiness. 

Although there are multiple proprietary credit scoring methodologies used across the bureaus, the FICO score is the dominant standard for assessing credit risk in the U.S.. 

The Digital Revolution

Early computers were large and expensive, but their potential for transforming data-driven workflows was obvious. With steady improvement in processing capabilities, computers could process complex data operations faster and more consistently than human beings.

The invention of electronic data interchange (EDI) in the 1960s offered a massive upgrade for how lenders and underwriters standardized and shared information. 

It was now possible for credit reports to be built, maintained, and shared in a consistent, reliable way. 

Electronic data systems helped improve the integrity of credit reporting because they revealed inconsistencies, but early credit reporting practices were still plagued by bias, discrimination, and other errors. 

Advanced Analytics and Real-Time Data

The internet was invented in the 1960s, but it wasn’t until the 2000s that it began disrupting the way banks handled consumer data and underwriting. 

Today, organizations are using real-time data, artificial intelligence (AI), and machine learning (ML) to improve their underwriting automation. Cloud computing, advanced analytics, and APIs have unlocked a world where data can move instantaneously and assessments can happen in the blink of an eye. 

Emerging Underwriting Technology

The latest improvements in underwriting tech aren’t universally applied by lenders.

Underwriting AI has changed so rapidly that many creditors and institutions are using “modern” underwriting models that pale in comparison to what AI-driven techniques can achieve. The pace of change has caught them off guard and left them at the mercy of early adopters

In the pursuit of better, faster underwriting models, a number of factors are shaping the next era of risk assessment and credit analysis. 

Alternative Data Sources

Credit reporting agencies only collect and process certain consumer data, painting incomplete pictures of each person’s financial health.

As a result, businesses have relied on alternative records such as rental history, utility payments, cell phone plans, and social media activity. This approach can also fill in gaps for people with minimal credit history. 

The trouble with alternative data is that it’s rarely standardized, leading to incomplete and outdated information. It also doesn’t carry as much weight as conventional risk and credit factors. 

ML- and AI-Driven Credit Scoring

One of the primary benefits of AI-driven analysis is that it can digest massive troves of data and look for meaningful relationships that human analysts would miss. It can also be trained on specific datasets, allowing for deeper insights into a specific demographic or customer profile. 

There’s lots of hype around what AI is capable of, so it’s smart for companies to maintain a healthy skepticism.

That said, AI is, by definition, a highly advanced statistical model that can learn and self-optimize. That makes it a perfect fit for the statistics-heavy practice of underwriting. 

Automated Decision-Making

Automated underwriting offers benefits and risks, depending on how you implement it. It’s best to think of automated underwriting as a preliminary filter that reduces the application review burden on your staff. 

You still need to ensure that you’re feeding the most complete and up-to-date data to your automated underwriting tool, but it’s a great way to increase throughput and the quality of your risk assessments. 

Cybersecurity

The sophistication and volume of cyber-attacks have only increased with time—aided by the advent of AI and the digital vulnerability of most consumers.

That means that lenders should be highly vigilant, implementing cybersecurity defenses such as multi-factor authentication and robust identity verification. 

The best cybersecurity measures also use AI to analyze transactions for anomalies and signs of fraud. Lenders need to fight fire with fire if they’re going to limit cybercrime risks.

Advanced Bank Verification™: Underwriting’s Game Changer

The next wave of improvements in underwriting will be driven by Advanced Bank VerificationTM (ABV), a process that enables financial institutions and companies to quickly and securely verify a consumer’s financial information. But that’s just the beginning.

At the same starting point as instant bank verification (IBV), ABV can analyze that data near-instantaneously and provide a comprehensive picture of the applicant’s financial health and history. Using AI, ABV can produce highly accurate risk assessments based on alternative data sources, including non-standard income and debt payments that usually get missed by conventional analysis. 

It can also strengthen your anti-fraud efforts by validating a consumer’s identity in real time, which reduces application review times and helps you flag emerging forms of fraud before they can cause lasting damage. 

ABV is the best way to harmonize conventional data, alternative data, and spending patterns into a robust risk assessment for your underwriting team. 

Underwriting Technology Can Transform Your Operation. If You Let It.

Most lenders and creditors have a high confidence in their underwriting standards. They’ve used best practices to grow their business up to this point.

So why fix what isn’t broken?

That’s a fair assessment, but it’s exactly the sort of attitude that opens the door for more bold, innovative companies to steal market share. 

Just because the existing approach to underwriting is “good” doesn’t mean that it couldn’t be much, much better. There’s never been a time in history when new technology was developing as quickly as it is now. The pace of change is orders of magnitude greater than it was just 25 years ago. 

By embracing Bankuity’s ABV, your company can stay at the cutting edge of underwriting technology and secure your future against the toughest competition. 

Smart decisions start with insights, get our expertise in your inbox.

You Might Also Be Interested In