3 Technology Gaps Facing Financial Institutions
Each year financial institutions spend billions of dollars on technology. This investment is driven by several factors, including evolving business needs, regulatory mandates, and findings from regulators or auditors. Despite this vast investment, there are still three fundamental areas in which there are significant and serious gaps in the technology infrastructure supporting basic banking functions in accounting and risk management for loan portfolios.
Closing these three gaps through an architecture-based approach allows banks to tackle current and future challenges better, faster, and cheaper than ever before.
1. Tackling the divide between loan servicing systems and the general ledger
Servicing systems cover the operational lifecycle of a loan and address questions including: what’s the customer balance of the loan, how much is the payment, what portion of the payment is applied to principal and interest, and so forth. The “cash view” of the servicing system is crucial to lending institutions, and existing software has performed well in this space for years. In fact, some organizations continue to use systems purchased when Star Wars was still a trilogy!
The general ledger covers the “accounting view” -- recognition of interest income, bank balance of loan, reserves held, and charge-off activity. The “accounting view” of this system is also critical to the institution.
Up until a few years ago, the “cash view” and the “accounting view” of the general ledger were closely connected, and banks used the servicing system to perform many of the accounting calculations. Outside of the general reserve (FAS 5), it was rare that the “accounting view” required inputs from other departments of the organization, such as the risk or credit functions.
In recent years the cash view and the accounting view have diverged. The financial crisis helped accelerate this divergence as it left banks with significant numbers of non-performing and modified loans on their balance sheets and a slew of new regulations that were written, at least in part, in response to the crisis. Servicing systems simply can no longer do what is necessary to account for loans, and banks have resorted to webs of complex spreadsheets and/or custom built IT solutions. In general, and in a best-case scenario, these solutions are expensive and risky. Addressing the challenges this way is especially difficult for cases where the servicing system has to be manually overridden so that the “cash view” -- the most important aspect of a loan -- remains correct.
2. Addressing the disconnect between the modelers and users of loan risk models
Historically, the creators of risk models have used tools for estimating models that range from spreadsheets for less complex models and statistical software programs -- such as SAS, Stata, R, or the like -- to estimate models using sophisticated econometric techniques. There are standard ways of thinking about and modeling credit risk for each asset class that are widely supported in academic literature. The groups that have traditionally built, owned, and run the risk models typically sit outside of the accounting and finance functions within the organization. The models are typically run in isolation and not integrated into other systems within the bank. Embedding the models on applications such as cashflow engines to generate forecasts of loan behavior has historically been the exception and not the rule, even for the very largest banks in the US.
A significant gap has emerged in recent years, as credit stress testing has become a fundamental risk management activity. It’s no longer enough to have models -- institutions need to integrate them into an enterprise-wide process.
This gap is detrimental for several reasons. Most importantly, it impedes an institution’s ability to perform analysis, manage risk, and provide inputs to key processes such as stress testing and accounting. This gap will become increasingly difficult to manage if, as expected, the FASB adopts the CECL impairment standard.
To effectively fix this issue, use an integrated model execution system that can host various types of models across all asset classes to forecast income and losses in an integrated and controlled manner. As of now, very few institutions have implemented this best-practice.
3. Filling the hole between relationship managers for commercial credits and risk management
Whereas mortgages are relatively homogeneous, commercial credits are frequently “deals.” These agreements often come loaded with complex terms and conditions.
Additionally, the credits are generally large enough that they are individually reviewed on a periodic basis. Moreover, non-performing loans are often considered to be troubled debt restructurings for accounting purposes, and the review has financial statement impacts. Combined with increased disclosure requirements and regulatory expectations, this technology gap is important.
It used to be that the information from underwriting and periodic loan reviews would be paper-based. Leveraging electronic document repositories is better, but still does not address the challenge of giving risk managers access to the information. In order to fill this gap, institutions need a system that gathers information in a structured manner and integrates the information into the institution’s data architecture and analytical framework. The system also needs to include a workflow component, allowing for secondary reviews of the loan manager’s reviews up the institution’s organizational structure.
Today’s business challenges are different from yesterday’s. Tomorrow’s will be different from today’s. Financial institutions have needed to adapt their processes, but evolving the technology has lagged. This has resulted in inefficiency (think: dollars and time wasted), control gaps (think: mistakes made in key processes) and people spending too much time processing and not enough time analyzing. The good news is these gaps can be bridged, but they will require institutions to adapt broader, architecture-based approaches.
John Lankenau is the head of valuation and accounting product solutions at Primatics Financial. He has extensive consulting and financial services industry experience, with an emphasis on complex loan systems integrating risk and finance. Mr. Lankenau is a notable thought ... View Full Bio
Tweet This