What credit unions need to know about accounting for the new credit loss standard.
The current expected credit loss standard—or CECL as it is more commonly called—will fundamentally change how U.S.-based financial institutions account for credit losses.
Before CECL, FIs used relatively simple analytical models to account for losses based on actual defaults or on specific events that indicated a loss. Set to take effect for credit unions Dec. 15, 2021, CECL will require CUs to calculate expected loss over the whole lifetime of a loan.
Below is a primer on this new accounting standard, including what you and your staff should be considering now to prepare for the changes ahead.
CECL involves three primary changes:
- The expected credit losses are to be estimated over the life of the loan. Estimation of life is a key decision point and demands careful analysis of the portfolio under consideration.
- CECL is forward-looking. It requires credit unions and other financial institutions to estimate losses for the foreseeable future, taking into account internal conditions related to the portfolio as well as such external factors as macroeconomic indicators. Therefore, more powerful loss forecasting models are needed with built-in adjustments, along with an easy-to-implement management overlay framework.
- Unlike U.S. Generally Accepted Accounting Principles, CECL requires loss provisioning for current exposure only. This creates a unique challenge for credit cards and other revolving products as FIs have to attribute future losses to the snapshot balance.
How Should CUs Prepare?
Although 2021 may seem like a long way off, credit unions should begin planning for CECL now so they are fully prepared when the changes take effect. Below are some suggested high-level milestones to keep in mind over the next couple years:
- Assess readiness and create roadmap by end of 2018
- Test different vendor solutions on portfolios and select best solution by mid-2019
- Prepare data and infrastructure for CECL by end of 2019
Learn what works from the practices and methodologies followed by large banks (that have an earlier deadline) by mid-2020
Credit unions should evaluate their internal resources and capabilities to decide whether they can prepare for CECL on their own or if they need outside help. Here are the major capabilities needed for a successful CECL implementation:
- Data: Granular loan-level information is required on balance, payments, origination risk score, loan to value, APR, etc. The length of history depends on the product type. Typically credit unions will need at least five years of historical data for most products.
- Dedicated team: Credit unions need a team to develop the forecasting approach, execute it on a regular basis and prepare reports for senior management and auditors.
- Subject matter expertise: To develop the models and create auditor-ready output, credit unions will require subject matter experts who are accustomed to working with regulations. Important roles include skilled data managers, experts in credit risk model development, compliance officers, financial accountants and project managers. Credit unions that lack any of the capabilities mentioned above may be well served to consider using outside help.
Credit unions choosing to set up their own internal process should start planning their CECL program sooner. Here are some steps they should plan to complete this year:
- Create the roadmap and project plan.
- Define roles and responsibilities for different teams.
- Review gaps in current data and infrastructure and create remediation plans to address them.
- Outline the analytical framework to forecast CECL losses, make adjustments, test the output and create final reports.
Even credit unions opting for a vendor solution should complete steps this year. These steps include:
- Assign a point of contact.
- Prepare a list of available vendor solutions.
- Meet different vendor teams, request demos and compare the features across different solutions in terms of data requirements, availability of different modeling approaches, output reports, documentation, user experience, etc.
Choosing the right vendor will depend on which capabilities a credit union is lacking. If a credit union does not have the required data but has the subject matter expertise and team to handle CECL, it may approach a credit bureau to gather their historical data.
If a credit union has the data but lacks personnel and subject matter experts to work on CECL, it may look to a firm that can augment its internal team. Or, it may look for vendor solutions that can use the credit union’s data to generate output. The former may be more suitable for larger credit unions while the latter may work better for mid-size and smaller credit unions.
What Impacts Will CECL Have?
CECL will mean major changes to the lending programs of all U.S. FIs—large and small. The effect on reserves should be similar (in terms of percent outstanding) on large and small credit unions, assuming the portfolio quality is similar. In most cases, CECL will result in an increase in reserves. The amount of increase will depend on the asset class (mortgage, credit cards, etc.) and portfolio quality. However, with a larger portfolio size, the amount of reserve will be higher for larger credit unions.
In addition, regulator expectations may vary depending on the size of the credit union. Larger credit unions will be expected to use more advanced modeling approaches, better documentation and better governance for CECL. The leading credit unions will be expected to create custom models for their portfolios. Smaller credit unions will likely be able to get by with less sophisticated approaches, industry models and less comprehensive documentation on the modeling and process.
Regardless of size, below are four key areas of business impact credit union leaders need to plan for.
1. Reserves and ability to lend
The new rules will most likely require credit unions to keep more reserves. Based on our industry experience, we predict the overall increase in reserves will be in the range of 20 to 40 percent. The increase will vary based on portfolio mix, product characteristics, etc.
Given the increase in reserves, the cost of credit will go up. Credit unions may decide to shift their lending policies toward low-risk consumers and may also offer a greater proportion of products with shorter terms to reduce reserves and increase profitability.
2. Mortgage and auto loans
The effect of CECL is expected to be greater on longer term loans like mortgages and auto loans. This is primarily because the losses are to be incorporated for the life of the loan. Under GAAP, on the other hand, typical allowance for loss on a 30-year mortgage loan is based on the next 12 months only. Although marginal losses will decrease with term of the loan, the total losses over life of the loan will be significantly more than 12 months of losses.
3. Internal operations
The new rules are primarily expected to impact five internal operations teams in the following ways: The data team will need to keep a more comprehensive, continuous and granular data repository relating to loans and their performance
- The risk management team will need to build CECL models or invest in the right industry solution for CECL
- The compliance team will need to understand nuances of CECL guidelines for attesting CECL reserves to avoid fines and other consequences of non-compliance
- The finance and accounting team will need to optimize cash flows to fund additional reserve requirements
- The product and marketing team can be expected to design shorter-term products and create associated go-to-market strategies In addition, credit unions with a diversified portfolio of lending products may need to set up a CECL project management team to coordinate with stakeholders across all lending products for alignment and efficient adoption of CECL.
Given that CECL is a new regulation, credit unions will be subjected to increased scrutiny from regulators in the initial years. Regulators will typically evaluate the governance process around CECL programs, including modeling approach, engagement of business owners, qualitative factors, documentation, etc.
Manish Jain is vice president for EXL, New York. He has worked in the field of data analytics since 2007, primarily for banking and financial firms in the marketing and risk disciplines. Manish has rich experience across the entire model life-cycle, from building and governance to monitoring and tracking.