One headline combines two different processes
Net subtract recoveries on previously charged-off loans from gross charge-offs. The Federal Reserve’s published quarterly rate uses net charge-offs divided by average loans and annualizes the result. The measure is useful, but today’s defaults and today’s collections may come from different customers and origination periods. An improving net rate does not establish that current borrowers are under less pressure. [1]
For finance teams, recoveries affect cash and reported credit performance. For servicing teams, they reflect the resources required to resolve old accounts. For customers, a settlement or sustainable repayment arrangement can matter more than a portfolio ratio. The OCC’s lending handbook supplies the portfolio-management context; the operating examples below are analysis, not prescribed collection targets. [2]
A hypothetical quarter
Assume a portfolio has $1 billion of average loans. In the first quarter it records $20 million of gross and $5 million of recoveries, producing $15 million of net charge-offs. Annualized mechanically, the net rate is 6%. In the next quarter gross charge-offs rise to $24 million while recoveries rise to $10 million, leaving $14 million net, or 5.6% on the same average balance.
The reported net rate improves even though new losses increase 20%. These figures are hypothetical and ignore day-count refinements. The point is that the second quarter’s recoveries may reflect accounts charged off years earlier. They can represent genuine economic value without establishing that recently originated loans are performing better. A sound review shows gross losses, recoveries and net losses separately.
Recovery curves explain the timing
A recovery curve tracks cumulative collections after , usually grouped by charge-off and months since charge-off. It helps distinguish a larger stock of collectible old accounts from better collection effectiveness. A portfolio with unusually heavy prior losses may generate strong current recoveries simply because more balances are available to work.
Compare like cohorts at comparable ages and account for changes in product, balance size, customer circumstances and collection channels. A twelve-month recovery percentage for one vintage should not be compared with a three-month percentage for another. Also distinguish gross cash collected from net economic recovery after agency fees, legal expense, servicing and other directly attributable costs.
Debt sales and operational changes
Selling charged-off accounts can accelerate cash receipt and change the reported recovery pattern. That can be economically sensible, but the sale price must be compared with expected net collections and risk under continued ownership. A one-time portfolio sale should not automatically be extrapolated into recurring monthly recovery performance.
Collection policy changes also affect timing. A revised settlement offer may produce more immediate cash at the expense of later collections; a new agency may initially work the easiest accounts. Measure the full curve and the mix of accounts assigned. Treat cash acceleration, higher lifetime recovery and reduced operating cost as separate potential benefits rather than adding them together without checking for overlap.
Connections to underwriting and the allowance
Gross and migration are often more direct signals about recently deteriorating credit than recoveries from old accounts. Compare new origination , payment behavior and alongside the net loss series. If early delinquency worsens while recoveries improve, management should investigate the underwriting change rather than conclude that the recovery team has neutralized future risk.
Expected recovery assumptions also affect lifetime-loss estimates. A favorable recent collection period may justify review, but it is not enough to assume a permanent improvement across all accounts. Economic conditions, legal processes, customer ability to pay and collection capacity can change together. Assumptions should reflect the relevant portfolio and horizon, with documented sensitivity to less favorable outcomes.
Controls and cost
Recommended controls reconcile recovery cash to account histories and previously charged-off balances, identify reversals and distinguish principal from other amounts where relevant to reporting. A reporting error that classifies fees or unrelated receipts as principal recovery can make credit performance look stronger. Source-to-ledger reconciliation is therefore part of credit-risk analysis, not only an accounting task.
Manage agencies and collection vendors using both financial and conduct measures. A lower fee can be offset by poor documentation, customer harm or ineffective dispute handling. Review complaints, settlements, documentation access and sampled account treatment alongside collection yield. Strong recovery economics should come from a sustainable process whose costs and risks are understood.
Gross receipts are not net economic recovery
Hypothetical example: collecting $1 million at $200,000 of direct expense produces $800,000 before other costs. A new approach collecting $1.2 million at $450,000 of expense produces $750,000. Gross recoveries rise 20%, yet net cash contribution falls $50,000. This is an economic comparison; it does not redefine the regulatory net numerator or assume collection expense is reported as a reduction of recoveries.
Timing matters as well. A discounted debt sale generates cash sooner but transfers later collection opportunities. An internal strategy may yield more over several years while consuming staff, technology and funding today. Compare discounted cash after relevant expenses on similar account pools, and distinguish one-time portfolio sales from recurring collection performance.
Resolution quality can change future workload
Analysis: a repayment arrangement that repeatedly fails can increase contacts and apparent activity without creating durable resolution. Track re-default after settlements or arrangements, repeat disputes, processing errors and the time required to correct an account. Segment outcomes by account age and available payment capacity rather than interpreting more contacts as better servicing.
These measures also help investors interpret recovery gains. Higher collections accompanied by fewer reopened cases are different from gains purchased through heavy operating expense. They do not, on their own, demonstrate compliant treatment; applicable collection and consumer-protection requirements remain separate obligations.
Read cash improvement and credit improvement separately
A convincing recovery improvement persists in comparable account cohorts after costs and does not rely on a temporary backlog liquidation. A convincing improvement in new lending also requires better gross losses, migration or mature- results. Present both findings when they differ.
The useful question is which process improved: originating loans, servicing live accounts, resolving old losses, or monetizing an inventory of charged-off debt. Each creates different expectations for future earnings and customer outcomes.
Sources
- Federal Reserve: Charge-off and delinquency rate methodology; reviewed September 29, 2026Official releaseBack to text: ↑
- OCC: Credit Card Lending, Comptroller’s Handbook; current posted edition reviewed September 29, 2026Official source · PDFBack to text: ↑