With aggregate household debt climbing the last 24 consecutive quarters to a record $13.5 trillion and the increasing likelihood of a softening economy, financial institutions should reassess the risk management practises in place for consumer lending. Additionally, with the impending deadline for implementation of the current expected credit loss standard (CECL), a closer look at lending practises can result in more accurate estimates of expected losses.
Specifically, we recommend that risk functions focus on early problem loan identification via heightened loan monitoring practises. This blog post focuses on three areas that can help risk managers enhance credit risk management practises over consumer portfolios:
- Leveraging analytics for enhanced portfolio monitoring
- Policy exception tracking, and
- Collections processes and policies
Leveraging Analytics for Enhanced Portfolio Monitoring
Due to the homogenous nature of consumer portfolios, financial institutions utilise portfolio-level data analysis in conjunction with the assessment of credit risk management processes and controls at the individual loan level to monitor credit risk. Institutions can improve on this practise by leveraging more advanced portfolio and process analytics, in addition to the usual loan-level criteria information (income and liquidity verification, credit score, repayment capacity, etc.).
The use of portfolio analytics is instrumental in the early detection of problem credits and/or portfolios, based on common consumer credit metrics. Strategic use of portfolio analytics further enables compliance with internal and external audit requirements driven by credit policy (e.g., annual or semi-annual review for low-risk customers).
When conducting portfolio analytics, some best practises include:
- Perform a “postmortem” analysis on delinquencies, modifications, cured credits, charge-offs, downgrades, non-performing loans and reasons for borrower hardship.
- Consider statistical analysis tools to analyse drivers of portfolio performance and behavior (e.g., payment history, utilisation, liquidity, etc.).
- Utilise data to evaluate processes for managing borrowers with financial difficulties.
- Assess portfolio concentrations: industry/profession, geography, new relationship managers and underwriters, loan approvers, and product mix (e.g., credit card, auto loan, etc.).
- Leverage the institution’s monitoring tools and capabilities such as flag reports to perform behavioral analysis on borrowers.
Other recommendations include:
- Perform loan system data integrity testing to ensure the accuracy of information used for portfolio monitoring.
- Include key risk indicators in portfolio analyses that inform of the severity and direction of credit risk and include these factors in formal risk assessments.
- Utilise adequate historical data in benchmarking to assess actual results, as compared to only using projections or quantitative benchmarks. Data should be segmented into smaller sub-populations for ongoing monitoring to better ascertain the risks within the portfolio.
Additionally, internal audit’s review of portfolio analytics will also provide insights into changes in risks and should be utilised as a tool to engage the lines of business and loan review function in discussions about how these risks are being monitored or remediated.
Policy Exception Tracking
An increased number of policy exceptions within a loan portfolio is often a key indicator of rising credit risk. The Office of the Comptroller of the Currency (OCC) notes in its Spring 2019 Semiannual Risk Perspective that while banks have generally maintained moderate underwriting practises, they have loosened underwriting and shown higher tolerances for policy exceptions. For credit card programmes, heightened competition and the rise of non-traditional lenders has encouraged relaxed underwriting standards and aggressive solicitation programmes, which can increase credit risk.
Key elements of policy exception tracking include:
- Well-defined policies – Clearly delineated, measurable underwriting criteria (minimum FICO score, debt-to-income thresholds, etc.) are the foundation for effective policy exception monitoring and ensure consistency in risk acceptance across the organisation.
- A system for identifying and tracking policy exceptions – Policy exceptions should be well-documented, included in approval considerations and accompanied by a discussion of the mitigating factors that warrant the exception.
- Monitoring and reporting – The number and exposure of new loans with policy exceptions should be proactively managed and reported to the appropriate levels of management.
For well-defined policy exception protocols, we suggest the following three main categories of exceptions:
- Credit policy (loan-to-value, debt-to-income, pricing)
- Credit scoring – high- and low-side overrides. (High-side override means borrower is declined despite meeting initial underwriting criteria. Low-side override means borrower is approved despite not meeting the defined criteria.)
- Documentation (loan application, transaction information, collateral perfection/valuation)
Collections Processes and Policies
Successful recovery of bad debt depends on the quality of the consumer debt collection unit, which should utilise skilled staff and follow sound policies that are consistent with applicable laws/regulations. We recommend the following best practises:
- Have clearly defined and comprehensive collections policies that address the full scope of collections activities and emphasise the following: authority levels, roles and responsibilities, collector compensation and collector monitoring. These policies should also address plans to identify, recruit, hire and train the requisite resources in a timely manner.
- Establish a playbook based on defined metrics which includes policy levers to drive collection efforts in the event of changes in the environment.
- Use collection practises like re-aging, extensions, deferrals, renewals and rewrites prudently as they may distort a portfolio’s performance or prove ineffective in managing credit loss.
- Evaluate collection reporting to determine adherence to policy and whether specific issues are being tracked/resolved based on the results of an independent review.
- Establish a clear feedback loop from the collections process to bank strategy discussions and modeling teams.
- Develop behavioral scores utilising historical statistical data to facilitate credit decisioning on collection accounts.
The process for establishing loss estimates normally consists of the following common elements, which are also relevant under CECL:
- Identify the factors that may affect the loss estimate.
- Accumulate sufficient and reliable data on which to base the estimate.
- Develop assumptions that represent management’s judgment of the most likely circumstances and events.
- Determine through appropriate governance processes the estimate based on the assumptions and other relevant factors.
- Determine that the accounting estimate is presented in conformity with applicable accounting principles and that disclosures are adequate.
As numerous economic indicators point towards an impending downturn in the credit cycle, institutions should make a concerted effort to further refine and heighten existing loan monitoring practises. While the best practises outlined above may not be exhaustive, they should provide institutions better information to manage consumer credit risk, minimise losses, and report the allowance for loan losses under CECL.