The Social Fabric of Finance

The Social Fabric of Finance
This post was published on the now-closed HuffPost Contributor platform. Contributors control their own work and posted freely to our site. If you need to flag this entry as abusive, send us an email.

Isolated Lending Finds a Friend in Social Networking
2015-11-23-1448313807-2659881-SocialFabricofFinance.png

For decades the lending industry has tirelessly pursued the perfect equation to determine a borrower's likelihood to repay a loan. Armed with Big Data and the algorithms to parse it, today's lenders claim to have found the key - but they are mistaken. As advanced and refined as their tools may be, the model these lenders employ is stricken with a fundamental flaw: the evaluation of borrowers in isolation.

For a true representation of a borrower's creditworthiness, we must also consider input from the community that surrounds them, the social fabric that in many ways defines each of us. These signals provide a far more accurate picture of a person's character than other proxies in use today, and with the rise of social networks, these signals can now be digitized at scale while also retaining objectivity.

To move the lending system forward into the 21st century, the lending industry must take advantage of the unprecedented level of digital connectedness exhibited by modern borrowers providing the infrastructure to digitize character reference and factor into borrower evaluation. We must take into consideration the social fabric of finance.

The Static State of Lending

As it stands, up to two thirds of US borrowers are paying too much for loans and are generally being underserved by financial institutions today.

How could this be possible? When lenders toute their mastery of Big Data and state-of-the-art selective algorithms to ensure borrowers get the most affordable, lowest interest loan possible - each one using a proprietary, secret mix of variables unlike any of the other - it would appear risk evaluation has been optimized to perfection.

The truth is, however, that no matter what process a traditional lender might advertise, the entire industry largely employs the same metric, commonly referred to as the FICO score. Created in the 1980's, this aggregate score is a snapshot of a person's creditworthiness based on the sum of their past financial decisions. As the saying goes, the best indicator of future behavior is past behavior.

In many ways, these advanced techniques are more effective than earlier methods.

The algorithms currently in use by FICO can sift vast mountains of data, scrutinizing even the most minute signals to judge an individual's propensity to repay a loan. Today's lenders tend to focus on decision models that incorporate hundreds or thousands of variables reflecting an individual borrower's past actions. By and large, these evaluations are largely based on a person's use of capital. For example, overdue payments, the number of credit lines opened and amount borrowed are all considered. However, in recent years, the calculation has advanced exponentially.

Innovations, in what is sometimes called credit analytics, have provided new ways for lenders to extract meaning from data sets. Even seemingly innocuous behavior, like the use of capital letters in a loan application, is being used to assess a prospective borrower's character, and therefore creditworthiness of today's borrowers.

Did you type an incomplete sentence on your loan application? How about scrolling down to the fine print and spending the time to read it? As inconsequential as these actions may seem to a person's creditworthiness, they have an effect on some lender's decisions

The logic is simple: With more data points available, the model becomes more accurate. True, but it misses the mark. The greatest limitation of the traditional loan decision making processes of today is not a lack of data about a borrower, rather the inherent isolation by which lending decisions are made.

There is a more effective way to set interest rates.

For a true representation of a borrower's creditworthiness, we must not only consider the borrower in isolation but also the trove of input from the community surrounding the borrower - their personal social network.

Networks Effects + Loans

For decades, there has been a push to leverage the power of computation in order to objectify the character of an individual, and to determine a more accurate measure of creditworthiness. Big Data was heralded as the solution we've all been waiting for, but while innovation in this space created efficiencies in some ways, the picture is still incomplete.

As an industry, finding the perfect equation can only become a reality by ending the isolation in which lending decisions are currently made. We must make evaluation a social endeavor.

This is not speculation. There are real-word examples clearly illustrating this method's effectiveness over others: community banks.

On a local-level, community banks have demonstrated how known social connections can influence the creditworthiness of an individual. When a potential borrower applies for a community bank loan, that bank then collects information about their social connections within the community by speaking with employers, friends and neighbors to gauge the character of the borrower. The result: community banks have a below industry average rate of default.

According to a report by the Mercatus Center at George Mason University, community banks have first-hand knowledge of their customers that provides them useful information for sound lending decisions. As a consequence, the loans made by community banks tend to default at lower rates than those made by larger financial institutions. The rate of loans in default for the first quarter of 2013 on loans secured by one to four family residential properties was 3.47 percent for banks with less than $1 billion, compared with a 10.42 percent default rate for banks with more than $1 billion in assets. Additionally, FDIC Chief Economist Rich Brown states, between 1993 and 2006 community banks had lower loan losses on average every year compared to larger institution.

Community banks offer a viable solution to the isolation problem. They understand that a person's relationships with others is one of the richest data sets in terms of discerning character. In addition, they realize that this type of data is stable over time. A lifelong relationship cannot be made overnight; it takes time to build and develop and therefore, represents a more accurate signal.

In contrast, factors contributing to modern credit analysis are becoming increasingly transient and speculative. The result is volatility.

Community-driven lending also promotes accountability. Individuals are accountable for not only their own financial reputation, but also those who lent support to them. Borrowers know a default reflects poorly both on themselves and their supporters, and because reputations are on the line, social participation creates a more responsible and involved borrower. This too has been proven time and time again by the community bank model.

In a study conducted by several financial researchers, they investigated whether the "ruralness" of small banks and small businesses had an influence on loan default rates. During the investigation, "ruralness" was defined as a tight knit social circle, which created a high stock of social capital in the surrounding area. Through economic modeling and data analysis, the researchers discovered rural community bank loans for businesses defaulted substantially less often than other loans issued by banks in larger cities. Also, rural community banks had a competitive advantage when lending to borrowers with limited or opaque credit history. The results were attributed to community banks' localized focus, and ability to foster intimate customer relationships.

Digitizing the Character Reference

While community banks have succeeded in outperforming the market on borrower evaluation, there's just one problem - and it's a big one - community banks cannot scale. But with every major problem comes the opportunity for a major solution.

In recent decades, banks have grown astronomically large - and as the scale has increased, the model for acquiring informative social data has broken down. In moving towards the digitized world we live in today, lenders lost the knowledge and ability to access this communal data, instead opting for an efficient, technologically-driven approach.

Tremendous growth of larger banking institutions has led to the degradation of the community banking model, both in terms of the number of charters and asset holdings. According to a study conducted by Working Papers in association with researchers from Rutgers University, more than 90 percent of U.S. banks are small community banks (with less than $1 billion in assets); however, approximately 90 percent of U.S. banking assets are held by larger banking institutions.

This tremendous shift in asset holdings has led to a historical consolidation of community banks. According to comments made by FDIC Chief Economist Rich Brown, from 1985 to 2010 individual charters for community banks have decreased from 18,000 to 7,700, of which 34 percent merged with different banking members and 24 percent consolidated within the same banking organization.

The rise in consolidations and mergers have negated many of the benefits of community banks. As these banks merge and grow larger, the relationship with the surrounding community can become disconnected. The valuable personal relationship between bank and borrower created at a community level takes a back seat to the bank's overall growth strategy. In turn, best practices from larger institutions are adopted such as big data credit analysis.

Until recently, there has been no way to gather or interpret the rich trove of data living within the personal networks surrounding borrowers. Now, the rise of online social networks has made it possible to explore a remarkably accurate digital representation of person's true character for the first time. This has paved the way for the digitization of the character reference, the skill community banks have mastered on the local-level.

On the macro-level, online social communication networks and the massive adoption of smartphones have led to an unprecedented level of connectivity in the world. People today are connected to each other via apps, email, text messages, address books, online social networks and beyond. This has created a unique opportunity to rethink our fundamental approach to lending.

The Social Fabric of Finance

With up to two thirds of US borrowers being penalty-priced, underserved, or poorly served, it's time to move the lending system forward and begin utilizing the unprecedented level of digital connectedness exhibited by modern borrowers to remove the inefficiencies and lack of data undermining the lending system today.

With the rise of social media platforms, lending is on the cusp of a revolution once again. Just as big-data efforts in the 80's provided new ways to decision loans, the rise of social networking and the ubiquitous nature of connectivity finally make it possible for the lending industry to increase inclusion and the access to capital for all borrowers.

The key is waiting within the social fabric of finance.

Popular in the Community

Close

What's Hot