Capital One's June Loss-Rate Decline Reflects Broad Credit Strength and Positive Vintage Trends

What explains the June loss rate decline, and how are the consumer and Capital One customers performing (especially with recent vintages)?

1) Why the June loss rate declined (what management pointed to)

• There was no one-off “sale” or similar mechanical event cited for June’s loss-rate improvement. In the domestic card context, management explicitly said “there is nothing to call out in the June domestic card charge off rate.” ¹
• The primary operating backdrop was consistently strong credit performance, with delinquencies moving about as expected (seasonality) and charge-offs described as “amazing.” Management noted that the June delinquencies “moved in line with seasonality” and that June was “still… great strength” even though it was the first month that did not beat seasonality. ¹
• Charge-off behavior in the quarter supports the conclusion that credit quality remained strong month-to-month, not that credit conditions suddenly improved due to a one-time item. In the domestic card segment summary, Capital One reported the domestic card charge-off rate was 4.71% in Q2, down 39 bps sequentially and down 54 bps year over year, with the delinquency rate 3.39% at quarter end, down 31 bps from the linked quarter and down 21 bps from a year ago. ²

Bottom line from the excerpts: the June loss-rate improvement appears to be consistent with a broader pattern of strong credit metrics (especially delinquencies and charge-offs) rather than a discrete June-specific transaction or adjustment. ¹²

2) Consumer performance: macro resilience plus improving customer metrics

Management characterized the consumer as resilient and supported that view with multiple macro indicators:

• Unemployment in June was lower than in February; jobless claims were low; job creation had rebounded; and consumer spending remained strong. ¹
• While some inflation-linked pressure existed earlier, June showed a turn back toward positive real wage growth. Management said real wage growth was negative in April and May (YoY) but returned to “positive… ever so slightly in June” as inflation ticked down. ³

They also tied consumer resilience to customer-level leading indicators inside their own book:

• In domestic card, credit metrics improved year over year in the quarter. ¹
• Payment rates were “meaningfully above pre pandemic levels across all… customer segments,” which they described as a healthy sign of credit quality (even if it can slow growth). ³
• Spend levels continued to show “health” driven by both account growth and spend per customer. ³
• Revolve rates stabilized near pre-pandemic levels for major products and segments. ³

Bottom line: the consumer story in these excerpts is not “deterioration avoided,” but rather continued strength—macro resilience plus multiple leading-leaning metrics (delinquencies, payment/revolve/spend) consistent with a better loss environment. ¹³

3) Capital One customers: strength across the business, including leading indicators

Domestic card + overall portfolio credit

• Management emphasized that credit continues to come in “very strong,” and highlighted delinquencies as their most important indicator. ¹
• They stated that charge-offs were “amazing” in that June/quarter context, paired with the delinquency/seasonality comment. ¹
• They also said the credit performance was improving year over year across domestic card. ³

Auto (as part of the same customer credit picture)

Management stated:

• Auto credit metrics were strong and they saw improvements in the quarter. ³
• They also reported auto charge-off rate of 1.43%, down 21 bps sequentially (and up 18 bps year over year due to mix shifting back toward pre-pandemic subprime levels). ⁴

Bottom line: the excerpted narrative is that credit performance strength is broad (card and auto) and supported by multiple internal leading indicators, consistent with the decline in loss-rate outcomes. ¹³⁴

4) How recent vintages are performing (what’s strong vs what’s still pressured)

The question specifically asks about recent vintages and whether they are performing “in line with expectations.” The excerpts contain two relevant vintage dynamics: one positive (legacy Capital One) and one that still carries over a prior-cycle headwind (Discover).

A) Legacy Capital One vintages (24 and 25): performing better than 22/23

Management was direct on performance by origination vintage:

• For legacy Capital One (and also Discover) they said they are seeing “24 and 2025 originations coming in better than 2022 and 2023.” ⁵
• They added that these 24/25 performances are “a bit below pre-pandemic levels” (i.e., still strong but not the very best reference period). ⁵
• They linked this to confidence in the marketing/origination spend, stating this outcome is “one of the things that gives us the confidence to lean into our originations spend… and… marketing.” ⁵

Interpretation: the “recent vintages” for legacy Capital One are described as better than earlier vintages (22/23) and thus consistent with (or better than) the company’s expectations for credit quality progression. ⁵

B) Discover recent vintages: growth brownout, but credit strength is the tradeoff

For Discover, management described a different vintage/maturation story:

• They explained that after Discover’s credit expansion in 2022 and 2023, Discover dialed back origination programs and credit line management toward the end of 2023 and largely sustained those pullbacks. ⁶
• As a result, “more recent smaller vintages mature” and the Discover portfolio “has been and continues to face some headwinds to growth.” ⁶
• However, the key qualification: the flip side is “strong credit performance” that is “playing through the system.” ⁶
• In the Q2 segment overview, they stated legacy Discover card loans shrank 1.5% year over year, which they said was “in line with our expectations for the temporary brownout of Discover loan growth.” ⁷

They also gave more detail on what’s behind the Discover brownout:

• Capital One had been trimming the margins of Discover’s credit policy in areas where they were less comfortable about resiliency, specifically mentioning high balance revolvers, and then connected that to the growth headwinds. ⁶

Interpretation: recent Discover vintages appear to be pressured on loan growth / outstandings, consistent with a deliberate tightening and integration-driven policy normalization, while credit quality remains strong—so “recent vintages performing” is nuanced by portfolio:

• Legacy Capital One (recent originations) = better than 22/23. ⁵
• Discover (more recent smaller vintages) = growth headwinds, but strong credit performance is expected/confirmed. ⁶⁷

5) Direct synthesis: what explains the June loss-rate decline and the “vintage” question

Putting the excerpts together:

• June’s loss-rate decline is best explained by ongoing credit strength (delinquencies around seasonality and charge-offs described as “amazing,” with no one-off June domestic card charge-off “call out”). ¹
• That strength is consistent with a broader set of consumer resilience and customer leading indicators (payment rates above pre-pandemic levels, stable revolve rates near pre-pandemic levels, spend health). ¹³
• For recent vintages, the company’s messaging splits:
  • Legacy Capital One originations (24/25) are coming in better than 22/23, supporting the view that recent vintages are performing at least as expected from a credit outcomes perspective. ⁵
  • Discover’s recent vintages are tied to the company’s “brownout” narrative—loan growth is pressured and outstandings are down (expected), but credit performance is strong, meaning recent vintages are not showing the kind of deterioration that would negate the loss-rate improvement trend. ⁶⁷

Net answer: June’s loss-rate decline appears to reflect stable-to-improving credit quality and strong charge-off/delinquency dynamics, while recent vintages are performing well for legacy Capital One and Discover’s more recent vintages are pressured on growth but supported by strong credit performance, consistent with management’s expectations. ¹⁵⁶⁷