bg-arrow-down icon-arrow-up icon-back-to-top icon-linkedin icon-menu icon-search icon-twitter logo-white slider-arrow-left-gray slider-arrow-left slider-arrow-right-gray slider-arrow-right

Banks Struggle With New Reserve Rules

Banks are grappling with a new rule that changes the way they account for losses, a move that is causing significant practical problems and may also have wide-reaching implications for loan pricing and overall product strategy.

The issue, which will be one of the biggest challenges facing banks in the coming year, is complicated by a lack of complete and historic loan data at
many institutions.

Given this shortfall, banks will need to start sharing data about the historic performance of their loan portfolios with each other if they are to be prepared for the new rule.

The change in GAAP standards, known as CECL (Current Expected Credit Loss), requires U.S. banks and credit unions to set aside reserves for each loan commensurate with the total future credit losses expected throughout the life of the loan. The move is aimed at reducing earnings volatility by better anticipating future problems. It may also result in bigger reserves and lower profits.

The new standard, which takes effect beginning Dec. 31, 2019, is creating intense work for banks (finance, credit, risk, and data departments) that will have to plow through loan origination data, ongoing servicing data, and ultimate loss and charge-off data. While most banks have done a better job collecting loss data in the years since the financial crisis, many challenges remain.

After years of a benign credit environment, many banks just don’t have a good sense of potential losses. That is particularly the case in some specific loan portfolios, such as risk measures at origination for CRE loans.

Additionally, a flurry of merger activity among smaller institutions means that some loan portfolios have incomplete data. Banks that have at least five or six different systems that contain the required information are bound to run into data-mapping issues.

By sharing loan-performance data, banks can get a better sense of performance when they model for the lifetime of their loans.

CECL is a significant change from the previous standard, called the Incurred Credit Loss method, in which banks set aside reserves for losses they expect to occur within the next 12 months. Such reserves usually only are set aside once an asset already experiences significant credit deterioration to the point where collection becomes doubtful.

The accounting change is partly aimed at addressing some of the fallout from the financial crisis when investors in bank stocks lamented the lack of foresight in bank balance sheets. Losses realized from defaults in loan portfolios far exceeded the reserves banks set aside for such scenarios.

The rules require banks to make “life of loan” reserve forecasts. This means all future expected losses must be reserved for at the time of origination, no matter how remote the probability of loss may be. The loss estimates must be based on “reasonable and supportable forecasts” and forecasts are expected to include forward-looking views of the economic environment in the bank’s primary markets.

It is expected that those macroeconomic forecasts need to extend at least one year, but as many as three years may be ideal. FASB’s guidelines indicate the macro forecasts should be “reasonable and supportable”, meaning that they are based on analysis, relevant to the bank’s market and can be predicted with a sense of accuracy.

FASB’s guidelines also state that banks can then revert to their historical loss rates to estimate longer-term losses since CECL requires loss estimates that go out to the entire life of the loan — which could be as much as 30 years for a mortgage.

The banking industry has expressed concern about the new rule. The Wall Street Journal reported earlier this year that executives at 18 U.S. regional banks asked Treasury Secretary Steven Mnuchin to conduct an analysis of the long-term economic effects of the rule. Some federal lawmakers are also worried about the rule, which could potentially cut into bank profits by forcing companies to set aside more for future losses.

The American Bankers Association has said that it is concerned about how banks will implement CECL, especially the burden it could place on community banks. The demands are particularly challenging for smaller institutions: a study by Pacific Coast Bankers’ Bank in California found that only 11% of chief financial officers at financial institutions with less than $10 billion in assets believed they are prepared for the CECL transition.

CECL’s reach goes far beyond the reserve estimates and their effect on the Allowance for Loan and Lease Losses (ALLL). Depending on the bank’s loan mix, it could experience a material change in its reserves, and most likely that change would be an increase. If the economic forecasts are reasonable and accurate, many banks should experience a reduction in ALLL volatility (and thus earnings volatility); however, less robust forecasts may lead to higher ALLL and earnings fluctuations.

Moreover, CECL has the potential to fundamentally change the economics of offering certain loan products. For example, longer-dated assets such as fixed-rate residential mortgages may realize large increases in reserves, thereby reducing their profitability. This, in turn, could lead to fundamental changes in loan pricing, and even in bank and product strategy.

The clock is ticking.

In order to be ready to produce updated financial statements by the end of 2019, most banks want to do a full year of parallel runs alongside current reporting, which means that models need to be in place, tested, and validated by the end of 2018. To meet that deadline, models need to be developed ideally by 3Q-4Q 2018, meaning the data needs to be in order by the end of 2Q 2018.

Data analysis efforts are underway at many banks, but the timeline is tight. Will your bank be ready?

Novantas is exploring a cooperative effort that would address CECL challenges by developing a loan-data consortium to help banks round out their data completeness.

The consortium would also help banks to develop their mandated loss-forecast models.

The Novantas loan-data service will gather large volumes of loan-level historical data from member banks across the country in an effort to provide them with a solid foundation on which they can build their CECL loss models.

Data will be transformed, cleaned, and standardized into a single format in line with Novantas’ data model. Anonymization ensures confidentiality of each member bank’s portfolio.

Novantas will also provide advanced benchmarking of a bank’s portfolio against the rest of the membership to enable member banks to understand how they compare to their peers.

For more information, contact Steve Wiggins at

For more information, contact Novantas Marketing

+1 (212) 953-4444

Related Materials


Sending the Right Signals: Industry Advancements in Behavioral Life Measurement

The prospect of rising interest rates and more competitive pricing is challenging one of the banking industry’s traditional ways to value deposits.


Crystal Ball: Banking on 2018

As we head into 2018, there are a few certainties for the banking industry. While we don’t have a crystal ball, we know a lot of smart folks who spend time thinking about these issues.


Lenders Need to Seek New Skills from Commercial Bankers

The commercial bank executive of tomorrow will look a lot like a quick-thinking technology executive or a nimble start-up CEO.