How One Bank Puts Agile Management Techniques Into Action

When David Mansfield took the reins as CEO of Provident Bancorp six years ago, he could see that a change was needed, and that required new thinking.

“We were a typical community bank trying to be everything to everybody,” says Mansfield. He transformed the $1.1 billion bank based in Amesbury, Massachusetts, into a “true commercial bank” to the small and mid-sized companies that form the “backbone” of the community.

We’re trying to offer products and services that are not commodities, where we can differentiate ourselves, add value and get paid for it,” says Mansfield. “The customer’s appreciative, because they’re getting a product or service that really isn’t available to [small and mid-sized companies]” — like specialty services usually offered by large regional and money-center banks to their corporate clients.

To accomplish that, he needed employees who weren’t afraid to shake things up. He also needed to develop a culture and tools that facilitated collaboration within the organization. To do this, he borrowed managerial techniques from the technology sector by adopting Lean and Agile techniques.

Teams within the bank using these methods identify how to improve processes and workflows. “We have had some really amazing success stories,” says Mansfield.

Lean management aims for continual, incremental improvement. Quick “daily huddles” in the bank help staff focus on the day. In these 15-minute standup meetings, employees provide a quick update about progress on key projects and share any obstacles they’re facing so these issues can be addressed.

Mansfield credits Lean methods for improving interdepartmental dynamics. “One of the major premises of Lean [is that] it’s all about the customer experience, and we truly believe within this organization that everybody has a customer,” he says. Loan officers and branch staff directly interact with the customer, but support staff have a customer, too: their colleagues serving the customer. “What I love about our IT group is, they believe that wouldn’t happen unless they serve their customer, which is that group of people.”

Provident Bancorp still incorporates Lean thinking, but started shifting to Agile techniques late last year, upon hiring Joy Curth as senior information officer. Curth’s experience includes a stint in application development at Intuit, and she understands Agile methods. The principles of Lean and Agile are similar; both seek to create workflow efficiencies and promote iterative development.

Curth doesn’t have a banking background, which appealed to Mansfield. “We’re trying to do some different things, really leverage technology, and the traditional bank chief information officer just is not what I was looking for,” he says. As the bank weighs partnerships with technology companies, “she’s not only able to speak their language, but she’s able to recruit people to join her team [and] really professionalized our project management team” due to her Agile background.

Adoption of Agile has been project based, and the bank’s first project under the methodology was integrating ResX Warehouse Lending, a warehousing lending division that it acquired in January from $58.6 billion People’s United Financial, based in Bridgeport, Connecticut.

“Dave came to us and announced we were going to do an acquisition, and we were able to complete that project in [roughly] 8 weeks,” says Curth. “A whole acquisition of staff, technology, contracts — that was pretty expedited and showed that we were able to do that without a hitch.” The project’s success encouraged bank leaders to roll out the approach for most key projects.

“Even the bank we were doing the acquisition from [was] really impressed with our team,” says Mansfield. “We really drove it; it was an everyday meeting, what’s the status, how to keep things going.”

Agile is an ongoing journey that Mansfield believes represents the “next evolution” for project management at Provident. He’s a big reader, and one of his favorites is “Good to Great: Why Some Companies Make the Leap…And Others Don’t,” by Jim Collins.

“There’s a concept he uses: Shoot bullets first,” says Mansfield. Shooting bullets means pursuing attempts that represent a low risk and require minimal resources. If it works, you recalibrate and then “shoot the cannonball when you’re ready,” he says — using your company’s resources to make a big move based on those earlier, iterative attempts.

Another one that he calls a “gut check” on Lean techniques is “Jumpstart Your Service Revolution: Transform Your Company’s DNA and Thrive in an Age of Disruption,” by Thomas Schlick.

By adopting Lean and Agile techniques, Mansfield is creating a bank that differentiates itself in the market. Curth adds that employees enjoy working there. It’s what drew her to the bank. “When you implement this type of culture, your morale is high, and there really is an energy that is compelling and exciting,” says Curth.

Recommended Reading from David Mansfield, Provident Bancorp

Evolving Considerations in the CECL Countdown


CECL-7-23-19.pngExecutives gearing up for the transition to the new loan loss accounting standard need to understand their methodologies and be prepared to explain them.

Many banks are well underway in their transition to the current expected credit loss methodology, or CECL, and coming up with a preliminary allowance estimate under the new standard. CECL will require banks to book their allowance based on expected credit losses for the life of their assets, rather than when the loss has been incurred.

The standard goes into effect for some institutions in 2020, which is slightly more than six months away. To prepare, executives are reviewing their bank’s initial CECL allowance, beginning to operationalize their process and preparing the documentation around their decision-making and approach. As they do this, they will need to keep in mind the following key considerations:

Bankers will need time to review their bank’s preliminary results and make adjustments as appropriate. Banks may be surprised by their initial allowance adjustments under CECL. Some banks with shorter-term portfolios have disclosed that they expect a decrease of their allowance under CECL, compared to the incurred loss estimate.

Some firms may find that they do not have the data needed to segment assets at the level they initially intended or to use certain loan loss methodologies. These findings will require a bank to spend more time evaluating different options, such as identifying simpler methodologies or switching to a segmentation approach that is less granular.

These preliminary CECL results may take longer to analyze and understand. Executives will need to understand how the assumptions the bank made influence the allowance. These assumptions include the periods from which the bank gathered its historical loss information for each segment, the reasonable and supportable forecast period, the reversion period, its prepayment assumptions, the contractual life of its loans—and how these interact. Bankers need to leave enough time for their institutions to iterate through this process and become comfortable with their results.

Incorporate less material or non-mainline loan asset classes into the overall process. Many banks spent last year determining and analyzing various loan loss methodologies and how those approaches would potentially impact their larger and more material asset classes. They should now broaden their focus to include less material or non-mainline asset classes as well.

Banks may be able to use a simplified methodology for these assets, but they will still need to be integrated into the bank’s core CECL process to satisfy internal controls and management and financial reporting.

Own the model and calculations. Executives will need to support their methodology elections and model calculations. This means they will need to explain the data and detailed calculations they used to develop their bank’s CECL estimate. It includes documenting why they decided that certain models or methodologies were the most appropriate for their institution and for specific portfolios, how they came to agree upon their key assumptions and what internal processes they use to validate and monitor their model’s performance.

Auditors and regulators expect the same level of scrutiny from executives whether the bank uses an internally developed model, engages with a vendor or purchases peer data. Executives may need specialized resources or additional internal governance and oversight to aid this process.

Know the qualitative adjustments. Qualitative adjustments may shift in the transition from an incurred loss approach to an expected lifetime one. Executives will need a deep understanding of the bank’s portfolios and how their concentration of risk has changed over time. They will also need to have an in-depth knowledge of the models and calculations their bank uses to determine the CECL allowance, so they can understand which credit characteristics and macro-economic variables are contemplated in the models. This knowledge will inform the need for additional qualitative adjustments.

Anticipate stakeholder questions. CECL adoption will require most banks to take a one-time capital charge to adjust the allowance. Executives will need to explain this charge to internal and external stakeholders. Moving from a rate versus volume attribution to a more complex set of drivers of the allowance estimate, including the incorporation of forecasted conditions, will require the production of additional analytics to properly assess and report on the change. Executives will need ensure their bank has proper reporting framework and structure to produce analytics at the portfolio, segment and, ultimately, loan level.

Five Tips on Choosing the Right CECL Solutions for Community Banks


CECL-10-8-18.pngAny big accounting change—especially one as large as CECL (current expected credit loss)—is bound to cause some pain. But, there are ways to make sure your bank is not making the challenge bigger than it has to be. Here are five tips on selecting a calculation methodology that’s compatible with your institution.

1. Consider the complexity.
Banks can choose from several methodologies that range in complexity. The more complex the methodology, the more data needed—and the more inherent risk of error. Both the Financial Accounting Standards Board (FASB) and regulators have consistently indicated that complex CECL models aren’t required. Nevertheless, some community institutions seem to be choosing more complex methodologies over simpler solutions that can decrease cost and reduce risk.

Overall, the choice of a more complex methodology can impose additional costs and risks to a bank. If a community bank is going to use a complex methodology, it should go into it with clear understanding of the cost and risk involved.

2. Select your methodology first.
Regulatory agencies — including the Securities and Exchange Commission and the FASB — have continually discussed how Excel is an acceptable tool for fulfilling CECL requirements. As a general rule, the more complex the methodology, the more likely you’ll need new software. Industry participants are becoming aware that they can use, with some adapting, methodologies similar to those they use today. That means they can continue to use Excel.

CECL software does have its advantages. For example, there’s functionality for quickly disaggregating the portfolio to a finer degree and the ability to explore various methodologies, which could be beneficial. But, new software won’t eliminate all of the hard work of making estimates requiring a managerial decision.

3. Don’t panic about the reasonable and supportable forecast requirements.
The accounting standard provides a framework for incorporating a reasonable and supportable forecast. The standard doesn’t require fancy and sophisticated forecasting techniques with regression equations. Using charts with historical economic information compared to long-term trend lines can be a way to reasonably support a forecast. This framework is illustrated within the accounting standard and consists of comparing the general direction of two economic indicators (unemployment and real estate values) and using historical loss periods with similar directional trends as a basis for qualitative adjustments.

4. Start with what makes sense and add complexity as needed.
The more complex the methodology, the more historical, loan-level data will be required. Many institutions won’t have accurate and complete data from several years ago readily available. They could do a tremendous amount of work right now to obtain that historical data. A better solution might be for those institutions to start changing their processes for the current year, so that going forward, they’ll have the correct data. In the meantime, they can use a less complex methodology that’s acceptable to regulators, such as the weighted average remaining maturity that doesn’t require loan-level information.

5. Ignore the hype and do what’s right for your institution.
Much of the focus in the industry now is on the big banks that are closest to adoption. A big, complex institution will require a complex CECL solution, so much of the dialogue in the industry relates to those complex methodologies. But, what’s good for your bank? Much of the industry buzz around advanced methodologies and CECL software has little to do with the needs of community institutions. The adoption deadline for community banks, which is still a couple of years away and simpler than that for large banks, is not an argument for procrastination. Rather, it’s a reminder that community institutions can craft solutions appropriate to their own needs that are efficient, effective, and economical.

Some community banks are still not working on CECL with necessary diligence and speed. Others are introducing complexity that makes the process more difficult than it has to be. An approach that recognizes there’s work to do—but understanding it can be minimized—is the right CECL strategy for the large majority of banks.