What is CECL?
CECL, or current expected credit loss, is a new accounting standard that is changing how financial institutions account for expected credit losses. Following the financial crisis, much of the immediate focus of the Federal Reserve and other supervisory authorities was on recapitalizing institutions and guarding against systemic risk with an increased focus on stress testing as the preferred tool to protect the global economy from further erosion. But perhaps more important to the bottom line are the revolutionary changes to accounting standards that determine the appropriate level of balance sheet reserves for credit losses.
In June 2016, the Financial Accounting Standards Board (better known as FASB) published its Accounting Standards Update (ASU) on Financial Instruments – Credit Losses (Topic 326). It replaces the prior standards addressing the accounting for credit losses– commonly known as FAS-5 and FAS-114.
FASB's CECL standards apply to any institution issuing credit, including banks, savings institutions, credit unions and holding companies filing under GAAP accounting standards. The rules affect all entities holding financial assets and net investment in leases that are not accounted for at fair value through net income.
The effective dates for CECL are phased, based on institution type. The accounting standard started Dec. 15, 2019 for public business entities that are US SEC filers – and will become effective in January 2023 for credit unions and all other lenders. Once effective, CECL fundamentally changes how these companies account for credit losses in their allowance for loan and lease losses (ALLL).
While the US CECL standards deviate in a few significant ways from the international IFRS 9 standard published two years earlier, they share an important feature – the calculation of the expected credit loss is now computed over the life of the loan. CECL represents a significant change from the previous incurred loss model. Under the previous incurred-loss model, banks recognized losses when they had reached a probable threshold of loss. Many analysts have suggested the older method for computing expected credit losses drastically underrepresented impairments and the calculation of potential future losses.
Free white paper
In Tackle the New Complexity of IFRS 9 and CECL Standards you'll find the best practices for getting your CECL implementation right. This paper includes recommendations for an organizational structure, data management, model development and much more.
Why is CECL important?
The impact of FASB's CECL standards is significant. While there is much variation across reporting banks, initial filers have reported higher loss reserve levels. And given that increased credit loss provisions represent the largest impacts to balance sheets under stress scenarios, the follow-on effects of these accounting changes will likely ripple through future stress tests as well. But while macro-prudential stress tests have typically focused on the largest banks, CECL applies to all financial institutions.
CECL not only affects how banks calculate credit loss reserves, but also how organizations fundamentally manage their ALLL and organizational processes for both finance and risk management. The scope of these changes can be substantial depending on the complexity of the balance sheet. The changes required by CECL require a much deeper level of loss modeling, analysis and reporting than what has previously been required. And these changes are significant in terms of how financial institutions will need to manage risk and financial data, build their analytic platforms and share information between departments.
The changes required by CECL require a much deeper level of loss modeling, analysis and reporting than what has previously been required.
Understanding CECL’s challenges
CECL brings some significant changes, and challenges, from practices under previous accounting rules. As I mentioned earlier, one of the most significant changes is the move from an incurred loss to an expected credit loss accounting framework. Additionally, while current rules require an allowance for credit losses only expected to incur over the next 12 months, CECL removes the probable loss threshold and requires a lifetime credit loss allowance to be established on day one of each exposure. This will require changes to credit loss forecasting, which prompts changes to the underlying credit loss models and loss aggregations for setting the allowance for credit losses. There are also changes to the accounting treatments of purchased credits and assets held for sale, which will require new modeling for these types of exposures.
Aside from new, more complex expected credit loss models, CECL requires more complete and detailed data. Currently, ALLL requires historical data on credit losses to be maintained. But CECL necessitates additional granularity. Not only is detailed data across assets important, macro-level data and risk factors need to be analyzed to assess the impact of various scenarios on credit losses.
Another potentially significant challenge for banks and credit unions is related to how they forecast credit losses over the full life of the asset. While incurred loss methods generally apply forecasts over a horizon determined over a loss emergence period, the lifetime loss estimates under CECL relies on a combination of two components:
- A forecast projected over a reasonable and supportable forecast horizon.
- An extension beyond that horizon using historical averages over the remaining life of the exposure.
There is flexibility in how these components are linked, and it will be up to firms to decide the best method. And because CECL standards are principles-based, they are not prescriptive in how institutions address specific modeling challenges.
CECL also requires increased transparency in the application of assumptions and in the disclosures around the allowance estimate. Under the expected credit loss approach, any justifications will have to be more quantitative in nature. That is true for any adjustments made on the front end as well as those made post-processing. Management's selection of forecasts or model outcomes (which may be a result of iterative model runs) will need quantitative backing to justify their selection.
The level of disclosures under CECL increase substantially. This includes disclosures that allocate reserves by origination date and also increased transparency around the overall process and assumptions that led to specific levels of loss reserves. Add to that the importance of defending assumptions, methodology choices and any adjustments, and it becomes critical to have a robust system to capture and govern the process from beginning to end.
Lessons learned from IFRS 9
To put these CECL challenges in perspective, it might be useful to look at what progressed outside of the US. In the rest of the world, IFRS 9 is the global standard that was issued by the International Accounting Standards Board to address expected credit loss accounting. It was finalized and published in 2014, and it has been adopted by more than 140 countries across the globe. Like the CECL standard, it attempts to provide a more forward-looking accounting of credit impairment by moving from an incurred-loss approach to an expected-loss model that considers the magnitude and timing of potential losses.
IFRS 9 differs from CECL in that it uses a three-stage classification to assess the time horizon used for reserving. The first stage – for exposures performing as expected – uses a shorter, 12-month credit loss period. As the credit quality of an exposure deteriorates, it enters the second and third stages based on threshold rules and then, like FASB's CECL model, it requires a lifetime expected credit loss reserve.
Because IFRS 9 was implemented prior to CECL, US financial institutions have an opportunity to learn from their international counterparts. While implementations are unique in specifics due to each organization’s internal systems and processes, there are some common themes and practices that emerged:
Planning: Coordination of resources pays off
Many banks working toward IFRS 9 initially underestimated the amount of work involved. As a result, they started slowly with limited dedicated resources and then found that their timelines have been compressed and in some cases slipped. To combat this, many organizations threw more people at the problem. This was a tactical approach but was not sustainable or cost effective. For a more sustainable approach to CECL, institutions need to take an integrated and holistic view that incorporates people, processes and systems.
IFRS 9 banks also found that the necessary coordination and integration between risk and finance was more crucial than they anticipated. Because the required effort was larger than initially estimated, many institutions had to increase their budgets and staffing accordingly. In many cases, banks also realized that they need help. There were numerous cases where the domain knowledge was not deep enough or where the change management skills required to orchestrate the organizational and process changes were not strong enough.
Additionally, when planning for implementation, organizations needed to consider the overall sustainability of revised processes going forward. Are these processes robust? Maintainable? Defensible?
All of this suggests a need for US institutions to take a more detailed look at how they address CECL and proactively plan for investments in IT, programs and resources to avoid potential issues down the road.
Data: Begin gathering and preparing all your data as early as possible
Although data challenges can be closely tied to modeling challenges, there are some unique issues with data that banks working through IFRS 9 needed to address.
First, the requirements for additional data has presented challenges for a number of institutions. For example, cash flow modeling requires integration of data from both risk and finance to model losses and payment streams (cash flows). This data is frequently housed in different systems, has different data definitions, are often populated at different times and have varying levels of detail.
And different expertise is needed to cleanse the data and apply expert judgment to align or augment it, especially in preparing it for credit modeling. This has added to the length of time it takes to model credit losses and raised issues for reconciliations and the maintenance of audit trails.
An ongoing challenge is dealing with missing or incomplete data. For certain data gaps, institutions have sought out publicly available industry data or third-party data. This reliance on third-party data often creates issues with definitions, timing, granularity and data lineage. Another problem is that the accompanying documentation may not be sufficiently transparent. A further challenge with reliance on third-party data is that it can be difficult to justify its use or defend its appropriateness or relevance. For example, industry loss data may not reflect the risk demographics of an institution’s specific portfolio.
A consideration for CECL (but not required for IFRS 9), is that of vintage analysis. Even if vintage-level modeling isn’t prescribed, the reporting aspect requires that sufficiently granular data is maintained throughout the process. When reporting is more detailed than the models it relies on, problems could arise. There is a considerable amount of reporting required to meet the regulatory requirements, as well as the needs of investors and management.
One key insight from IFRS 9 is that early, timely analysis and documentation of data (including sources, quality, history and enrichments) is needed and often takes longer and involves more resources than expected. In addition, data availability and quality may also influence the types of models that are to be used.
Models: Streamlined and pragmatic models are better than ‘perfect’ ones
One of the early and often revisited decisions institutions face is to determine if they should use their existing models or if new models are needed. For example, some organizations initially looked at their point-in-time loss models used for stress testing to determine if the level of detail and the horizon would be acceptable. In some cases, the answer was “yes” and in some cases, “no.”
For loss estimation where no model exists, or for which the level of detail is insufficient, new models need to be developed. Some experimentation might be required as different model types, segmentation schemes, and assumptions are investigated and reviewed. Lifetime credit loss estimates can vary greatly based on changes to many parameters so early testing is needed to ensure stability in actual use.
Another decision that institutions faced is related to the level of sophistication of their models. Many IFRS 9 banks originally planned to start with highly sophisticated models. Over time, they realized the implications of this (long development cycles, data requirements, etc.) and became more pragmatic. Given the complexity and time required to review and implement sophisticated expected credit loss models, many banks moved to more streamlined models, and are constructing processes that can be improved over time.
A lesson learned here for US institutions still implementing CECL is that perfection should not be an initial goal. Institutions should start with simple models and introduce enhancements incrementally. This will allow firms to begin establishing a workflow and process to provide and review the potential impacts on ALLL from the various inputs and model types. An incremental approach will provide the flexibility to adapt to changes in interpretations and any future regulatory guidance that emerges.
Governance: Start building audit and oversight capabilities now
Although credit loss modeling is one of the most visible features of IFRS 9, banks that focused solely on the modeling aspects of the new accounting standard often ended up having to backtrack and retrofit their processes to establish proper controls. Governance over the entire process, including all aspects of data, models, integration and reporting is critical to a successful outcome.
Banks that have also built their processes in silos have struggled with assembling a comprehensive set of data for review. Having a holistic approach, with transparent and understandable processes, facilitates auditability, supports repeatability and reduces “key person” risks. And because reserves are a critical component of financial statements, the consequences of failure can be much more immediate and severe (consider the possibility of a restatement of financials versus a material risk assessment).
Given this need for transparency and coordination, US firms will need to plan for and build in robust capabilities for audit, appropriate governance and financial reporting compliance as well as create a framework to support expectations around model risk management.
Recommendations for CECL implementation
Financial institutions are recognizing, at least conceptually, that many of the components of the allowance process are also used in other functions, such as stress testing and capital planning. To use these components, institutions must be diligent in breaking down the silos that exist in current allowance and other regulatory and business processes and capitalize on a more integrated solution. Institutions will also want to avoid designing systems and processes that continually tax their available staff or jeopardize their deadlines each quarter.
Another consideration is that the CECL process is more computationally intensive than the current incurred credit loss method. Under the current standard, only a subset of loans (those that had crossed the probable threshold of loss) were modeled, while under the CECL standard, firms need to account for the contribution of losses from all loans. Because of this, institutions must plan for additional capacity of their model execution platforms. Efficiency is the key here. CECL provides a rich opportunity to review and possibly optimize model execution processes with an eye on how they could be done more efficiently.
Lenders that are looking for early success should focus on:
- Solutions with a modular, open design approach that are adaptable to the changing interpretations of the new standards.
- Systems and processes that support iterative development cycles with the ability to revise and upgrade individual model components as new models are tested and reviewed.
- An effective architecture and adaptable framework include a centralized model library, a common data platform, centralized workflow orchestration, dynamic reporting capabilities, audit support, and robust governance and controls.
CECL solutions that use a modular, open, adaptive architecture such as the SAS® Solution for CECL more readily fit the reality that banks and credit unions live in today. They provide for more efficient, dynamic and sustainable solutions. And given that CECL is non-prescriptive about the approach used it provides the most readily flexible approach to addressing the evolving accounting standards and other regulatory compliance requirements.
About the Author
Tom Kimner leads the Risk Marketing and Operations area within the Risk Research and Quantitative Solutions division at SAS. He is responsible for executing the overall marketing plan for risk management solutions as well as coordinating risk priorities and operations on a global basis. Prior to joining SAS, Kimner spent the bulk of his career at Fannie Mae in various senior management roles spearheading corporate initiatives to more effectively manage credit and financial risk. He also worked for a housing and finance regulatory agency and a Washington think tank. Kimner has testified before the Financial Services Committee of the US House of Representatives and regularly speaks at risk conferences and other SAS-hosted events.
- Article Risk data infrastructure: Staying afloat on the regulatory floodWhat are the challenges of a risk data infrastructure and how can they be addressed? Here's what you need to know to build an effective enterprise risk and finance reporting warehouse that will effectively address compliance requirements.
- Article Model risk management: Vital to regulatory and business sustainabilitySloppy model risk management can lead to failure to gain regulatory approval for capital plans, financial loss, damage to a bank's reputation and loss of shareholder value. Learn how to improve model risk management by establishing controls and guidelines to measure and address model risk at every stage of the life cycle.
- Article Risk data aggregation: Transparency, controls and governance are needed for data quality and reportingFinancial institutions’ data aggregation and reporting techniques and systems are receiving increased attention both internally and externally. Find out how to take a comprehensive approach to BCBS principles and risk data aggregation and management.
- Article Beyond IFRS 17 – what's next?IFRS 17 is not just a new accounting standard. Its fundamental objective is to provide transparency and insight to the insurance business while identifying strengths and areas for improvement. Learn how to keep a long-term vision and achieve broader business value beyond the immediate demands of IFRS 17.