A hundred years ago, community banks relied almost entirely on relationship banking. Lenders knew the borrower personally, which provided a unique customer experience, allowing each bank to carve out its own local niche. Not only was the borrower happy with the personalized experience, banks were also able to use the relationship to gauge and monitor the borrower’s credit worthiness. Those that were better at ‘sizing you up’ made better loans.  To this day, relationship banking is still the cornerstone of the community banking model and drives a community bank’s competitive advantage.

Yet times have changed. To preserve the edge afforded by relationship-focused banking, most community banks must find a way to operate in a world where data-driven models allow greater lender efficiency, more robust portfolio monitoring and a better understanding of credit risk. Whether it’s CECL adoption, underwriting, loan pricing, strategic planning or M&A, community banks must strive to operate at the intersection of big data and relationship banking.

Community banks need to strike the delicate balance of “knowing thy borrower” and implementing big data – they must move into a new way of operating or risk extinction. Such is the nature of a competitive marketplace. In this new era, the most successful community banks will be those that balance the scales. But how?

Much of the discussion around big data is focused on why we should embrace it. What I find so fascinating is that no one would disagree with the sentiment. However, the practical application of big data to specific use cases is entirely absent from much of the discourse. The monopolization of big data in banking has been altogether ignored. This is especially true for community banks.

Many FinTech firms that can leverage the structured bank data ecosystem aren’t sure if they’re competitive with or complementary to community banks – ultimately blocking any ‘data relief’ that could otherwise be available through partnerships. Big data firms are more than happy to sell their Data as a Service to the largest financial institutions for prices that are cost-prohibitive for the average community bank.

Small banks are left with hollow or vague recommendations to ‘hire data scientists’ or ‘pay for a database’ with little or no consideration of the budget constraints that community banks already contend with – much less consideration of how the information should be deployed. So naturally, to the large banks go the data spoils.

The picture I’ve painted may seem extreme, but we have witnessed massive industry consolidation that is expected to continue. Many community banks can no longer deliver the ROI of their larger counterparts, much less compete with firms in other industries for investment dollars.

I’m not saying that the data dilemma is the cause of industry consolidation, but you’d be hard pressed to say that it wasn’t one of the issues.  A community bank that has access to big data will be in a better position to outperform its peers. So how do we get there?

Perhaps the answer lies at the heart of the community bank business model: relationship banking.  But I’m not talking about relationships with customers. What if community banks turned to each other?

Community banks should consider openly sharing and mining each other’s data. In fact, the industry has been primed for this reality. Regulation – as tedious as it can be – has forced community banks to store and maintain their data in a ‘structured’ way (think FDIC Alert Files, Call Reports, etc.). This data is ripe for consumption, there just isn’t enough under any given bank’s roof.

Consider the following hypothetical scenarios:

Community Bank of Exampleton (CBE), a publicly traded $5 billion institution, has struggled with growth and is considering lowering loan rates to entice borrowers. The danger, of course, is that it lowers prices too much and becomes a price leader – significantly undercutting the market. This would drive significant growth but would destroy the bank’s Net Interest Margin, sending a negative signal to the market.

First State Bank of Exampleton (FSBE), which isn’t publicly traded and is smaller than CBE, feels confident about its pricing strategy but has significant CRE concentrations. FSBE management has recently learned that risk ratings drive much of their internal stress testing and that the subjectivity in the ratings (which are assigned by lending officers) is causing drastic fluctuations across otherwise similar loans. Regulators are beginning to question the validity of the stress test and put pressure on the bank to lower its concentration levels.

First National Bank of Exampleton (FNBE) is considering entering into the booming CRE market in the Exampleton MSA. However, it has limited experience in the area – having focused mostly on consumer mortgages. The bank already hired a CRE expert, but she isn’t quite familiar with the Exampleton market. Management is unsure how to structure its CRE loans to carve out a niche.

Fortunately for these three banks, they are participants in the local Community Bank Data Exchange, where all banks in Exampleton (55 total institutions) have agreed to contribute loan-level information in a standardized format. While they can’t see individual account-level information (the Community Bank Data Exchange guarantees anonymity and only sends back out aggregated information), they can see aggregated portfolio level information, as described below:

CBE queries the database for loans that share similar characteristics to its own portfolio and learns that it is currently priced 22 percent above market rates in its major products. CBE management – now knowing the true pricing story – decides to benchmark its pricing at 10 percent above market, resulting in significant growth in its client base and, ultimately, bottom line.

FSBE samples the Community Bank Data Exchange and learns that, while it has significant variations in its own risk rating on CRE loans, most similar loans in the database show a risk rating of three. Management instructs the lenders to perform a quick check against the database when risk ratings are assigned to ensure consistency across the bank.  With more robust risk-rating control driven by benchmarking against an open database, the stress test results become much less variable, and examiners conclude that FSBE management has a strong handle on the risk generated by its CRE concentration.

FNBE wants to know how loans are structured in the Community Bank Data Exchange. The bank learns that most lenders have been originating floating rate loans with a spread of 1.5 percent above Prime. Rather than follow the market, management makes the informed decision to try and price its loans off LIBOR with a spread of 2.5 percent - creating an attractive lending option that was otherwise less available in the market.

The time to get involved in a data-sharing initiative for community banks is now.

Breaking down the data dam will lead to a deluge of financial data that community banks can pair with their relationship banking approach to make smarter, faster decisions, ranging from the evaluation of credit risk, to loan pricing, to making strategic decisions about markets.  Big banks are already sharing information, and community banks should share as well.

Guy LeBlanc  can be reached at gleblanc@invictusgrp.com