The COVID-19 crisis has affected all people in every sector of society, including consumers who need credit and the lenders that provide it.
As economies shut down, payment moratoriums linger, and unemployment skyrockets, consumers will face increased challenges in making debt payments. For virtually any consumer-facing institution that has future receivables—such as credit card and telecommunications companies, retailers, universities, health systems, utilities, or retail banks—the difficulty in collecting will lead to a sharp increase in the number of customers requiring financial assistance. Moreover, our forecast of credit card losses in the United States between 2020 and 2022 predicts a substantial increase in delinquency—potentially more than three times the current levels. Similarly, mortgage losses in the next 12 months could be as high as 33 times today’s levels. When added to the specific circumstances of the crisis, these increased delinquency rates will overwhelm current credit risk strategies.
In the short term, risk leaders need to take emergency action to ensure continuity and compassion in customer assistance.
Beyond that, much work lies ahead to ensure that customers receive the support they need—and that lenders can continue cultivating relationships with their borrowers—while preserving shareholder value in the longer term. Sheltering in place could extend for several months and could permanently change working models and how people interact with each other and with digital channels. The pandemic-induced economic crisis has rendered existing credit risk segmentations useless, while current segmentation methodologies and segment-specific strategies are most likely just as ineffective.
Credit risk leaders have no time to lose—now is the time to formulate their longer-term response to the structural changes brought about, or accelerated, by the pandemic. That response starts by building the foundational capabilities needed to tackle their post-crisis reality. Risk teams must assess and reshape four core dimensions of delinquency management: segmentation, frontline operations (particularly those with remote workers), digital channels, and debt relief. Also, regardless of an institution’s level of sophistication, to successfully reshape its operations, risk teams need to build capabilities in agility, analytics, and infrastructure.
Institutions that fail to take action may quickly find their operations unable to properly and efficiently support their clients. Those that successfully build these capabilities for the long term to foster relationships with their customers and help them manage their obligations will be better equipped to reduce credit losses while helping their clients navigate the most volatile economic crisis in recent decades.
We strive to provide individuals with disabilities equal access to our website. If you would like information about this content we will be happy to work with you. Please email us at: McKinsey_Website_Accessibility@mckinsey.com
Building operational capabilities for the long term
Based on their organization’s size, level of sophistication, and appetite for change, risk leaders have to determine where and how to build operational capabilities in four key areas.
Currently, segmentation is done by analyzing historical data to classify the risk of customers. Some lenders use simplistic scorecard-based models, and nearly all use models that are several years old. Given the statistical noise generated from the coronavirus outbreak and related economic effects, these outdated models won’t be nearly as useful in predicting future behavior as they once were. Models created before the pandemic do not account for the differences in debtors’ income depending on the sector or region in which they work. For example, worldwide, a vast number of travel and tourism workers have lost their jobs, while those in transportation and logistics have suffered far fewer losses.
Such drastic changes in the environment raise the importance of learning segmentation models that quickly adapt and reassess the significant variables in the post-crisis setting. While sophisticated lenders are using standard analytics techniques, such as logistic regressions and decision trees, true machine-learning techniques will explore all possible combinations of segmentation and strategy and use frequent feedback to dynamically suggest the best policy for each customer.
Accurate segmentation is a crucial tool to optimize the effectiveness and efficiency of a lender’s operation, mainly when there are capacity constraints. Operations that do not yet have a structured segmentation approach based on value at risk should lay foundations to deploy segmentation for post-crisis scenarios as it will be critical to better allocate capacity. Additionally, customer-base segmentation can be used beyond capacity and risk optimization. It can also incorporate other elements of the operation traditionally handled through generic business rules such as channel and offer optimization and predelinquency actions.
To get the most from their segmentations, lenders could implement advanced analytic modeling that uses a mix of classification (that gives a probability for an event happening) and optimization (which is used to test multiple approaches to get an optimal result) to suggest a segmentation strategy mix that optimizes contact strategy, recoveries, cost, and customer experience.
Accurate segmentation is a crucial tool to optimize the effectiveness and efficiency of a lender’s operation, mainly when there are capacity constraints.
Given the next normal, segmentation strategy should be forward-looking. Historical data will help, but it won’t be as effective for predicting risk and behavior. Agility and the ability to continuously optimize segmentation strategy will be fundamental to success.
Managing the front line
As the global economic crisis induced by COVID-19 hits the personal finances of individuals, more borrowers will need to contact their lenders. The resulting increase of incoming customer communications will heighten the importance of properly managing traditional call-center employees who may continue to work from home even after COVID-19 quarantines have ended.
Most lenders already have capacity constraints in their call centers, whether those operations are internal or outsourced. Anticipating an increased volume of customer calls, leading chief financial and revenue officers are investing in capabilities and tools to analyze real-time (or near-to-real-time) information on their customer-service operations. These capabilities, combined with improving operational excellence, will be vital to get the most out of the limited existing capacity. To that end, retaining and investing in control desks and operation managers who can maintain high operational performance will be crucial to managing delinquency, particularly in high-touch, high-cost channels. Those professionals accomplish this goal by making data-based intraday adjustments such as changing the set break times of operators to later in the day to avoid them being away when contact rates are higher, thus reaching more borrowers.
The sudden shift to remote work may have started as a contingency response to the crisis, but adding or retaining remote workers after the crisis abates could help lenders increase their operational resilience while providing the benefit of added flexibility to employees. For many lenders, however, maintaining high efficiency levels with a remote workforce means acquiring a new set of management capabilities and facing a host of additional challenges. For instance, giving remote workers access to customer data creates a risk for cybercrime or other adverse behavior. These risks can be mitigated through greater activity monitoring. For example, tracking workstation screens and webcam images—while maintaining privacy—can ensure quality, productivity, and that a given job is performed only by the assigned, certified agent. Such technology opens up the opportunity for flexible work arrangements (in accordance with local regulation) that could further improve productivity and operational efficiency. In extreme cases, agents can have fully flexible schedules and be paid for the time they’re logged in and available for calls rather than for a rigid 9–5 workday.
By our estimates, lenders that keep 20 to 30 percent of their staff remote could save on infrastructure costs (perhaps by closing a physical office), which account for about 5 percent of the total agent cost—as well as reduce turnover and keep their best talent. Lenders that decide to permanently shift part of their workforce to remote work, however, will have to crack the code on how to adopt new best practices for performance management and coaching. In this new setting, supervisors and managers play even more important roles. Real-time information—primarily about how operators are spending their time—is fundamental to supporting daily performance conversations and offering insight on where they need to be coached. New remote-work tools can promote capability building, as well as facilitate daily huddles to discuss performance and metrics, one-on-one meetings to reinforce key teachings.
Adding digital channels
As consumers continue to demand more mobile and online solutions for conducting their business, many lenders have already made digital channels part of their strategy. However, the COVID-19 outbreak triggered an onslaught of payment deferrals, forcing many lenders to speed up their deployment of low-touch channels—such as robot calls, SMS, messaging apps, email, social media ads, and web portals. Additionally, using these channels to make proactive contact with customers early in the crisis can help reduce the number of accounts that become delinquent. Having a consistent end-to-end strategy to communicate with borrowers, from predelinquency stages to recovery, is critically essential for lenders to navigate this crisis.
To get the most from digital channels—and to ensure they effectively draw some volume from high-touch channels such as call centers—lenders must set an end-to-end multichannel strategy and create a consistent message for customers to ensure a seamless customer experience. This is particularly true and challenging when third-party providers are the ones making first contact with customers.
During the crisis, lenders can proactively use low-touch methods to remind not-yet-delinquent customers that their payment will soon be due. They can redirect or prompt those borrowers to visit self-serve channels, where they can complete questionnaires that give lenders a better understanding of borrower needs and current cash flow so lenders can adjust their payment schedules accordingly. Not only will this approach help lower the number of customers who become past due, but it also provides a bespoke customer experience that helps strengthen the relationship with those customers. Indeed, when the time comes that borrowers have cash available to make some payments, they are more likely to pay creditors with which they have a good rapport.
Lenders can use high-touch methods to contact high-value customers who might be struggling with their cash flows. A task force dedicated to specific accounts generally has a better understanding of a given situation, thus allowing them to negotiate more effectively. This deeper connection, in turn, can improve customer experience and lead to better collection rates. Consistency in messaging and support for high-value customers should go past traditional touchpoints for collections while incorporating recovery efforts. Given the volume of troubled borrowers amid the crisis—and the assumption that economies and cash flows will stabilize at some point (though the next normal remains to be seen)—an approach involving collections agencies and litigation may be inadequate or inefficient. Instead, lenders should consider creating a recovery team that extends debt relief and customizes efforts to help each high-value client meet obligations.
Making decisions on debt relief
Given the likely scale and depth of this economic crisis, numerous borrowers of all types may be willing to make regular payments but lack the financial means. Short- to medium-term relief will be essential to avoid their accounts becoming uncollectible.
Five practices can help lenders find the right level of relief:
- Maintaining the affordability of a loan. At a minimum, lenders can suspend associated fees and interest to shrink the size of a customer’s due payment. Avoiding a total loss is always preferable to earning marginal fees.
- Determining who is eligible for credit. Customers still need access to credit, even if they can’t service it. In such circumstances, lenders need specialized underwriters to tie credit to a customer’s projected income and future ability to service debt.
- Considering collateralized loans. For riskier customers, backing lending with guarantees or physical collateral boosts borrower accountability and provides lenders with
- Facilitating timely payments. Paying on time will be crucial, even if customers cannot make full payments. Lenders that regularly engage with customers, ideally through digital channels such as email or push notifications, will enjoy more regular payments and a greater chance of full payments resuming when the crisis has passed.
- Educating borrowers. Lenders should consider that borrowers’ debt priorities (such as credit cards, auto loans, and mortgages) may change over time. Those that educate and accommodate their customers on, for example, available government financial aid, can significantly improve outcomes.
Across all options for providing debt relief, the ability to tailor solutions to individual customers will need to be balanced by the associated operational complexity. Finding the sweet spot between offer flexibility and organizational demands, based on inherent capabilities and maturity of the organization, will be essential for success.
The functional capabilities lenders need
Despite periods of apparent stability, particularly in low-delinquency markets, best practices in delinquency management are constantly changing. The social and economic crisis brought by the outbreak of COVID-19 has accentuated these changes. Having an agile mindset and operating model, as well as the right underlying data analytics and infrastructure, will be critical for institutions to advance their operational capabilities at the necessary pace.
An agile mindset
The numerous structural changes in lending and delinquency management created or accelerated by the coronavirus crisis are complex and will require time to be properly calibrated. For example, refining customer segments, adjusting the channel strategy to best interact with them, and finding the best possible settlement offers will require several iterations of trial and error to achieve an optimal result.
Strategy and segmentation fine-tuning, A/B testing, pilots of new processes and channels, and continuous improvement are well-known best practices in customer service. But as lenders face fast-growing volumes of customer inquiries as well as the need to proactively contact borrowers before they become delinquent, these practices have become a core competence. To quickly test and implement operational improvements across all facets of delinquency management, lenders will have to adopt agile principles and methods:
- Assemble a multifunctional team to, for example, identify areas of underperformance and structure improvement tests.
- Conduct multiple A/B tests based on data, and use partial results to adjust quickly.
- Implement a structured and fast review process for high-value initiatives, such as investments in technology.
Lenders able to infuse these agile methods into their strategic discussions and customer service can quickly test and develop new approaches.
They will be able to espouse continuous improvement, optimize their operations for the next normal faster, and be better positioned to succeed in the long term.
Lenders able to infuse agile methods into their strategic discussions and customer service can quickly test and develop new approaches.
Data analytics and infrastructure
The evolution of advanced-analytics modeling capabilities will be central to supporting the speed and granularity of decisions on segmentation, debt-relief offers, and contact strategy. For instance, given the recent dramatic changes in borrower behaviors and an uncertain period of adjustment, customer segments will have to be modeled using machine-learning techniques that reasonably predict performance. Existing traditional modeling, on the other hand, uses a long series of historical data to shape and backtest risk segments, which can’t possibly account for changes caused by current events.
Finally, the growing demand for remote work requires that customers’ information and history of previous interactions be accessible, and agents must be able to capture and submit new information to deploy analytics models and adjust their operational parameters. For lenders to sustain remote work over the long term, they will need to make necessary adjustments to their front-end product and agent systems as well as to data storage, access, and security.
Meeting both data and analytical requirements takes a big toll on underlying infrastructure. More so than before, lenders will need to connect information from multiple channels, agencies, and portfolios. And they will need that information in near-real time to make decisions quickly and deploy models in production.
Because these operational changes must be done in a time of crisis, a cross-functional task force isolated from day-to-day business can serve as an agile-improvement cell. Depending on the size of the operation, task-force members might only be part-time. But it is key to have people from different areas of the organization—such as strategy, operations, risk, and the control desk—to structure operational changes and push improvements. A multidisciplinary task force also has a more robust end-to-end view on how to implement the changes while avoiding bottlenecks. Its mandate should give the team autonomy to make decisions without getting buried in a hierarchical chain within organizational silos.
Certainly, for most lenders, devising and implementing a long-term strategy for dealing with financially distressed customers will take time. With proper support and focus, most lenders can implement this in four to six months. But changed mindsets and continuous improvement should go on indefinitely. Indeed, lenders must start building the necessary foundational capabilities now to minimize the ill effects of the COVID-19 outbreak on their customers as well as their bottom lines.