By Alexander Flores, Debt Buyer & Triton Financial Solutions, LLC · Level 3: Advanced · Last updated: October 25, 2026
Three sets of numbers landed between August and October 2026 that matter to anyone pricing charged-off debt. The New York Fed’s Q2 2026 Household Debt and Credit Report showed card balances rising while the share of card debt that is 90 or more days late edged down from Q1, though it stayed above its year-earlier level. The Federal Reserve’s G.19 release for August showed revolving credit shrinking at a 4.2% annual rate. And the two largest public U.S. debt buyers reported very different second quarters: Encore Capital bought a record $372 million of U.S. portfolios, while PRA Group cut its U.S. buying by 40%.
This article is for experienced sellers, buyers and agency owners. It lines up the data by where it sits in the delinquency pipeline, shows what can and cannot be concluded from it, and closes with a practical playbook for Q4 2026 and early 2027. Every figure comes from the primary sources listed at the end, each checked on October 25, 2026. Where a number is our own calculation, we say so. Nothing here is legal, tax, accounting or investment advice.
Is charged-off debt supply rising or falling in late 2026?
Supply looks steady to slightly higher in the near term. Card balances grew 4.5% year over year, and the share 90+ days late stays near 13%. Bank card charge-offs held at 3.82% in Q2 2026, below 2024 levels.
The short version is that the pipeline is full but not overflowing. The NY Fed data shows $1.263 trillion of credit card balances at the end of June 2026, up from $1.209 trillion a year earlier (a 4.5% increase, our calculation from the NY Fed data file). Even if loss rates stay flat, a bigger balance base produces more charged-off dollars.
At the same time, the Fed’s charge-off rate for credit cards at commercial banks was 3.82% (seasonally adjusted) in both Q1 and Q2 2026, down from 4.19% in Q2 2025 and from 4.69% in Q3 2024. A flat rate on a larger base still means steady or rising dollar supply, which fits what Encore described as “ample portfolio supply” in the U.S.
Where does each data series sit in the pipeline?
Each series measures a different stage. Delinquency transitions lead, 90+ balances follow, bank charge-offs come next, and buyer purchases show what actually traded. Reading them in that order helps you estimate supply two to three quarters out.
Open-end credit like credit cards is generally charged off at 180 days past due, and closed-end loans at 120 days, under the federal banking agencies’ Uniform Retail Credit Classification and Account Management Policy (OCC Bulletin 2000-20). That rule creates a built-in lag. A card account that rolls into 90+ days late in one quarter usually reaches charge-off about one to two quarters later, if it does not cure.
| Pipeline stage | Data series | Latest reading | Earlier reading | What it tells a pricer |
|---|---|---|---|---|
| New early delinquency | NY Fed: card balances newly 30+ days late (Q2 2026) | 8.69% | 8.61% (Q1 2026) | Future inflow into collections queues |
| New serious delinquency | NY Fed: card balances newly 90+ days late (Q2 2026) | 6.97% | 6.93% (Q2 2025) | Accounts likely to charge off within about 1–2 quarters |
| Stock of serious delinquency | NY Fed: share of card balances 90+ days late (Q2 2026) | 12.92% | 12.27% (Q2 2025) | Size of the near-term charge-off queue |
| Realized loss | Fed: bank credit card charge-off rate, SA (Q2 2026) | 3.82% | 4.19% (Q2 2025) | Rate at which paper leaves bank books |
| Traded volume | Encore U.S. purchases (Q2 2026) | $372.3M | $317.3M (Q2 2025) | Large-buyer appetite in the U.S. spot and flow market |
| Traded volume | PRA U.S. purchases (Q2 2026) | $109.4M | $182.3M (Q2 2025) | Large-buyer return discipline |
Sources: NY Fed Quarterly Report on Household Debt and Credit, 2026:Q2 data file (released August 2026); Federal Reserve Charge-Off and Delinquency Rates, last updated August 25, 2026; Encore Capital Q2 2026 earnings release (August 5, 2026); PRA Group Q2 2026 Form 10-Q.
One note on the NY Fed numbers: they come from consumer credit reports (the NY Fed Consumer Credit Panel, based on Equifax data) and cover all lenders, not only banks. The Fed charge-off rate covers commercial banks only. The two will not match, and they are not meant to.
Why did revolving credit fall 4.2% in August 2026?
The G.19 release does not give a cause. It reports that revolving credit fell at a 4.2% annual rate in August, after rising 2.5% in July and 6.0% in June. August figures are preliminary and can be revised.
Here is what the Fed’s G.19 release of October 7, 2026 shows (seasonally adjusted):
| Month (2026) | Total consumer credit growth (annual rate) | Revolving growth (annual rate) | Nonrevolving growth (annual rate) | Revolving outstanding |
|---|---|---|---|---|
| June (revised) | 3.6% | 6.0% | 2.7% | $1,354.4B |
| July (revised) | 4.1% | 2.5% | 4.7% | $1,357.2B |
| August (preliminary) | 1.9% | -4.2% | 4.1% | $1,352.4B |
Source: Federal Reserve G.19 Consumer Credit, released October 7, 2026.
We will not guess at the reason. It could reflect paydowns, tighter lending, charge-offs removing balances, or noise in a single month. What matters for portfolio pricing is the direction over several months. If revolving balances keep shrinking into 2027, the supply of fresh card charge-offs in 2028 would be smaller than today’s balance base suggests. One month is not a trend, and the next G.19 release should be read before anyone changes a pricing model.
The same release shows the cost side for consumers. The average interest rate on all credit card accounts at commercial banks was 21.19% in August 2026, and 22.36% on accounts assessed interest. The 60-month new car loan rate was 7.54%. With the federal funds target range raised to 3.75%–4.00% on September 16, 2026, those rates are unlikely to fall soon. The FOMC meets again October 27–28, and we will not predict that outcome.
What do Encore and PRA’s Q2 2026 results say about pricing?
They point in two directions. Encore bought a record $372 million in the U.S. and said supply is ample. PRA cut U.S. buying 40% to $109 million while its U.S. Core purchase price multiple rose to 2.16x.
Both companies are public, so their filings are the best window into large-buyer behavior. Here are the Q2 2026 figures side by side:
| Metric (Q2 2026) | Encore Capital | PRA Group |
|---|---|---|
| Total portfolio purchases | $443.8M (up 21%) | $296.6M (down 14.4%) |
| U.S. purchases | $372.3M (record; $317.3M in Q2 2025) | $109.4M (down 40.0% from $182.3M) |
| Collections | $736.9M (up 13%, record) | $559M (up 4%) |
| Estimated remaining collections (ERC) | $10.18B (up 9%) | $8.9B (up 7%) |
| Forward flow disclosed | Not stated in release | $219M over next 12 months ($86M U.S.) |
| U.S. Core purchase price multiple | Not stated in release | 2.16x (up from Q1 2026) |
| 2026 purchase guidance | $1.4B–$1.5B | Not stated in release |
Sources: Encore Capital Q2 2026 earnings release, August 5, 2026; PRA Group Q2 2026 earnings release, August 6, 2026, and Form 10-Q for the period ended June 30, 2026.
A purchase price multiple (PPM) is the buyer’s expected total collections divided by the price paid. A higher PPM means the buyer paid less for each dollar it expects to collect. PRA’s U.S. Core PPM rising to 2.16x, while its U.S. volume fell, is consistent with a buyer that walked away from deals priced too high for its return targets. Encore’s record buying shows that other capital was willing to transact at the prices available.
For sellers, the lesson is not that one company is right. It is that large buyers are not one market. They run different return targets, funding costs and collection platforms, and they price the same file differently. That is the core reason a competitive, confidential process tends to matter more than the headline market trend.
How much did PRA pay per dollar of face value?
From PRA’s 10-Q, its Core purchases cost about 13.5 cents per dollar of face in Q2 2026, versus about 14.1 cents a year earlier. These are our calculations, blending regions, not a U.S. market price.
PRA’s 10-Q discloses Core face value of $1,983.8 million and a Core purchase price of $267.5 million for the quarter, compared with $2,269.4 million face and $319.6 million price a year earlier. Dividing price by face gives the blended figures above.
Use this number carefully. It mixes geographies, asset types and account ages. Fresh bank card paper and aged, previously worked accounts can trade many cents apart. It is useful as a directional check that large-buyer pricing did not jump in 2026. It is not a quote for any specific portfolio.
How should an advanced buyer turn these signals into a bid?
Start from projected collections, divide by your target PPM, then adjust for cost of funds and pipeline timing. The data above changes your inputs, such as liquidation curves and supply competition, not the basic formula.
A simple framework that many buyers use, shown here with example numbers only:
| Step | Input | Example (illustrative only) |
|---|---|---|
| 1. Project lifetime gross collections | Liquidation curve from your own or comparable static pools | 30 cents per dollar of face |
| 2. Choose a target PPM | Based on your cost of capital, collection costs and risk | 2.0x |
| 3. Maximum price before adjustments | Step 1 ÷ Step 2 | 15.0 cents |
| 4. Adjust for timing | Slower curves cost more when funding rates are higher | Lower the bid if cash comes later |
| 5. Adjust for file quality | Media availability, chain of title, data fields, state mix | Lower the bid for gaps or extra compliance work |
How the late-2026 data changes those inputs:
- Cost of funds. With the federal funds range at 3.75%–4.00% after September’s hike, leveraged buyers face higher carrying costs on slower-paying paper. That tends to favor portfolios with front-loaded collections. Encore highlighted a $1 billion refinancing in its Q2 release, a reminder that funding terms shape bid capacity.
- Consumer capacity. Card APRs above 21% and auto delinquency rising (5.49% of auto balances 90+ days late in Q2 2026, up from 4.99% a year earlier) suggest some consumers are stretched across several debts at once. Payment plans may need to be smaller and longer, which affects the shape of the liquidation curve.
- Competition for supply. If dollar supply stays ample, as Encore described, disciplined buyers can be selective. If revolving balances keep shrinking, competition for fresh paper may tighten in 2027–2028.
- Collections environment. The NY Fed reported that 4.88% of consumers had a third-party collection on their credit report in Q2 2026, with an average amount of $1,577. A year earlier the figures were 4.74% and $1,581.
What should sellers do with this data before year-end?
Sellers should time sales around clean files, not headlines. Prepare a complete data tape, confirm media and chain of title, and run a confidential, competitive process so different buyer return targets compete for the same file.
Practical steps for credit card, auto deficiency, installment, utility, telecom and other charged-off portfolios:
- Price against the buyer pool, not one buyer. As Encore and PRA show, two well-capitalized buyers can look at the same market and make opposite decisions. A wider, qualified pool improves the odds that your file meets someone’s return target.
- Fix the file before you go to market. Buyers discount for missing fields, unclear charge-off dates, gaps in account-level documents, and state-licensing questions. Our DebtHub resources cover data tape and media checklists.
- Review forward flow terms. If you sell on forward flow, check whether your agreement allows price reviews. PRA alone disclosed $219 million of forward flow commitments over the next 12 months, so flow remains a real channel. Pricing set in a different rate environment may no longer fit.
- Watch the auto pipeline. Rising auto 90+ delinquency means more auto deficiency balances could follow repossession and sale. Deficiency files need clean sale notices and deficiency calculations to price well.
- Know the next data dates. Encore reports Q3 2026 results on November 4, and PRA reports on November 5. Their purchase volumes and PPM commentary will show whether the Q2 split continued.
DebtMarket runs an NDA-first process with no public auction and a network of 200+ qualified buyers. When the file is complete, closings typically take 1-14 days. Sellers can start at /sellers/, and buyers can request access at /buyers/.
What are the limits of this analysis?
The data is national, lagged and partly preliminary. It cannot price a specific portfolio. August G.19 figures may be revised, Q3 earnings are not out yet, and the NY Fed switched its credit score model in 2026.
Specific limits to keep in mind:
- Timing. The NY Fed and Fed charge-off data are for Q2 2026 (through June). Q3 data had not been released when this was written.
- Score change. Starting with 2026:Q1, the NY Fed report uses VantageScore 4.0 for its credit score charts. Score-based comparisons with earlier reports need care. The delinquency and balance figures used here are not score-based.
- Coverage. Fed charge-off rates cover commercial banks only. Credit unions, finance companies and fintech lenders are outside that series, though they appear in the NY Fed and G.19 data.
- No secondary-market price index. There is no public, official index of charged-off debt prices. The PRA cents-per-dollar figure is a single company’s blended number.
- Company data. Encore and PRA describe their own buying, not the whole market.
Key takeaways
- Card balances are up 4.5% year over year, and about 12.9% of card balances are 90+ days late, so near-term charge-off dollar supply looks steady to rising.
- Bank card charge-off rates held at 3.82% in Q2 2026, below 2024 and 2025 levels.
- Revolving credit fell at a 4.2% annual rate in August 2026. One preliminary month is not a trend, but it is worth tracking for 2027–2028 supply.
- Large buyers split in Q2: Encore bought a record $372 million in the U.S., while PRA cut U.S. buying 40% and raised its U.S. Core PPM to 2.16x.
- Higher funding costs reward fast-paying, well-documented files. Sellers who prepare their files and run a competitive, confidential process are best placed whatever the macro data does next.
Related reading on DebtMarket News
- Our breakdown of the Fed’s Q2 2026 charge-off and delinquency rates
- What the September 2026 Fed rate hike means for collections and debt sales
- How fresh and aged charged-off paper price differently
Sources (each checked live on October 25, 2026)
- Federal Reserve Bank of New York, Quarterly Report on Household Debt and Credit, 2026:Q2 (released August 2026) and its data file.
- Federal Reserve Board, G.19 Consumer Credit, August 2026, released October 7, 2026.
- Federal Reserve Board, Charge-Off Rates, All Banks, Seasonally Adjusted and Delinquency Rates, All Banks, Seasonally Adjusted, last updated August 25, 2026.
- Federal Reserve Board, FOMC statement, September 16, 2026, and FOMC meeting calendar.
- Encore Capital Group, Second Quarter 2026 Financial Results, August 5, 2026 (SEC Form 8-K exhibit).
- PRA Group, Second Quarter 2026 Results, August 6, 2026, and Form 10-Q for the period ended June 30, 2026.
- Encore Capital Group, Q3 2026 results date announcement, October 6, 2026; PRA Group, Q3 2026 results date announcement, October 15, 2026.
- Office of the Comptroller of the Currency, OCC Bulletin 2000-20, Uniform Retail Credit Classification and Account Management Policy, June 2000.
This article is general market commentary for industry professionals. It is not legal, tax, accounting or investment advice. Consult your own advisors before buying or selling any portfolio.