20.94% Card APR Versus New Accounts Still 17% Below the 2022 Peak

The price of revolving credit and the pace of new revolving accounts no longer move together. Federal Reserve G.19 commercial-bank interest rates on credit-card plans print an all-accounts average of 20.94% in May 2026 (2026 Q2 survey month) — only 0.82 percentage points below the 21.76% peak in 2024 Q3, and still 6.43 points above the 14.51% trough in 2021 Q4. Over the same post-hike window, Philadelphia Fed FR Y-14M large-bank originations show quarterly new accounts at 17.12 million in 2026 Q1, 17.9% below the 20.86 million cycle high in 2022 Q3. Calendar-year openings fell from 79.2 million in 2022 to 68.9 million in 2024 and only edged to 69.6 million in 2025. The dashboard above pairs those series: dual-axis path, APR×accounts scatter, annual ladders, underwriting mix, and regime bars.

What each series actually measures

G.19 terms of credit are survey averages of commercial-bank interest rates on credit-card plans. The Board publishes two cuts that matter for this question. All accounts (FRED `TERMCBCCALLNS`) blends interest-bearing and non-interest-bearing balances; it is the cleaner “sticker APR” path for the industry. Accounts assessed interest (`TERMCBCCINTNS`) covers only revolvers who paid finance charges — it sits higher, peaking at 23.37% in 2024 Q3 and printing 22.15% in May 2026. Neither series is a consumer-facing offer distribution; both are bank-reported averages.

New-account volume here is not a household census. The Philadelphia Fed’s Large Bank Credit Card Data aggregates FR Y-14M filings from BHCs, IHCs, and covered savings-and-loan holding companies with $100 billion or more in assets that meet card-portfolio materiality thresholds. `RCCCONUMACT` counts new accounts; `RCCCOORG` sums original credit limits in dollars; `RCCCOACTPCTSCORELT660` tracks the share of those new accounts with scores below 660. The panel is large-bank heavy by design. The New York Fed Consumer Credit Panel supplies stock context — credit-card balances stood near $1.26 trillion in 2026 Q2 — but does not replace the Y-14M origination pulse for this dual-axis cut.

The APR ascent that did not reverse

From 2019 through early 2022, all-accounts APRs lived in a narrow band around 14.5–15.1%. The break came with the policy-rate cycle. By 2022 Q3 the all-accounts print was 16.27%; by 2022 Q4 it was 19.07%; by 2023 Q4 it was 21.47%. The peak arrived in 2024 Q3 at 21.76%. Since then the series has drifted only modestly: 21.47% (2024 Q4), 21.37% / 21.16% / 21.39% / 20.97% across 2025, then 21.00% and 20.94% in the first two 2026 prints. That is a plateau, not a unwind. Assessed-interest APRs tell the same story with a higher intercept: they crossed 22% in 2023 Q2, topped 23.37% in 2024 Q3, and remain above 22% in the latest print.

For households that revolve, the economic meaning is blunt. A 6-plus point lift on an interest-bearing balance compounds into larger minimum payments and slower principal paydown even when nominal purchase volume holds up. CFPB market reporting has separately documented that general-purpose purchase volume kept climbing into 2024 even as APRs rose — a reminder that price and activity are not the same object.

New accounts: peak in 2022, cooler thereafter

Y-14M new-account counts rebuilt quickly after the 2020 Q2 collapse (9.2 million). By 2021 Q4 they were 20.58 million; 2022 Q2–Q4 printed 20.35, 20.86, and 20.57 million — the cycle crest. 2023 stayed hot: four quarters between 18.17 and 19.95 million, summing to 77.5 million for the year. The break is 2024. First-quarter openings fell to 16.80 million (−7.4% year over year and −17% quarter over quarter, in line with the Philadelphia Fed’s own narrative), and the rest of 2024 never recovered the 19–21 million band. Full-year 2024 openings were 68.9 million, roughly 13% below 2022.

2025 did not restore the prior pace. The year opened at a soft 15.40 million in Q1 — the weakest post-2021 quarter — then recovered into the mid-to-high teens, finishing at 18.93 million in Q4. The annual total (69.6 million) is essentially flat versus 2024 and still about 12% below the 2022 peak year. 2026 Q1 printed 17.12 million, still 17.9% under the 2022 Q3 high-water mark. Credit-limit dollars tell a parallel story: peak $114.8 billion in 2023 Q2, softest recent print $97.1 billion in 2025 Q1, latest $105.5 billion in 2026 Q1.

Underwriting tightened even as sticker APRs stayed high

The composition of who got approved shifted with the rate path. The share of large-bank new accounts with scores below 660 peaked near 25.2% in 2021 Q4, then ground down to a trough of 15.6% in 2024 Q3. Implied average original limits rose from roughly $4.5k in late 2021 to about $6.1k by 2026 Q1. Banks did not simply slam the door; they redirected capacity toward higher-score borrowers with larger assigned lines while the subprime share of new accounts compressed. That mix shift helps explain why dollar limit originations fell less than account counts from the 2023 peak: fewer thin-file accounts, larger limits on the accounts that clear.

A modest reopening in the sub-660 share — back to 19.2% by 2026 Q1 — has not restored 2022-style volume. The underwriting panel in the dashboard shows the two lines diverging: sub-660 share down, average limit up, through the heart of the high-APR regime.

Three regimes on one chart

Collapsing the quarterly path into regimes clarifies the answer to the brief. In a pre-hike plateau (2019 Q1–2022 Q1), average all-accounts APR was about 14.8% and average quarterly openings about 17.0 million. During the APR ascent (2022 Q2–2023 Q4), average APR jumped to roughly 19.2% while openings stayed near 19.9 million — price rose first; volume had not yet cooled. In the high-APR, slower-openings window (2024 Q1–2026 Q1), average APR sits near 21.3% while average quarterly openings fall to about 17.3 million. The scatter view makes the same point geometrically: elevated-APR quarters cluster left-of-peak on the account axis.

What the dual path means for households and issuers

For issuers, a high APR plateau with cooler origination volume is a portfolio-management equilibrium, not a paradox. Existing revolvers carry finance-charge yields near cycle highs; new-account acquisition can be paced to credit performance and funding costs without chasing 2022 unit volume. For households, the combination is less friendly: fewer new lines at the margin, especially for lower-score applicants during the trough, while the cost of carrying balances on accounts that already exist remains elevated. NY Fed CCP stock still shows a large revolving book — $1.26 trillion in 2026 Q2 — so the APR plateau bites a wide stock even when the flow of new accounts slows.

Seasonality still matters. First quarters routinely soft-print after holiday openings, and 2025 Q1 / 2026 Q1 fit that pattern. The annual totals remove most of that noise: 2024 and 2025 are both well below 2022–2023. That is the durable signal.

Caveats and what this desk is not claiming

Several limits apply. First, Y-14M is a large-bank panel, not the universe of credit unions, smaller banks, or fintech-issued cards outside the reporting perimeter; national CFPB origination charts can diverge in level even when they rhyme in direction. Second, G.19 APRs are commercial-bank averages, not the distribution of offers a shopper sees on comparison sites, and they do not isolate promotional 0% periods. Third, account counts and credit-limit dollars answer different questions — a stable dollar total with fewer accounts is a larger average line, not identical credit supply. Fourth, 2026 Q2 openings are not yet in the Y-14M print used here; the dashboard carries Q1 account and limit levels beside the newer G.19 APR for visual continuity and flags that carry explicitly. Fifth, panel revisions can rewrite history when firms enter or exit FR Y-14M reporting.

This desk also does not equate slower openings with weaker household demand alone. Underwriting, marketing appetite, competitive rewards economics, and charge-off expectations all gate origination. The joint fact remains: commercial-bank card APRs stayed elevated above 20.9% while large-bank new-account pace cooled from its 2022 peak and has not recovered. That is the answer to the core question.

Sources

• Federal Reserve Board, G.19 Consumer Credit (terms of credit at commercial banks)
• FRED series TERMCBCCALLNS and TERMCBCCINTNS
• Federal Reserve Bank of Philadelphia, Large Bank Credit Card and Mortgage Data (FR Y-14M); FRED RCCCONUMACT, RCCCOORG, RCCCOACTPCTSCORELT660
• Federal Reserve Bank of New York, Quarterly Report on Household Debt and Credit (Consumer Credit Panel)
• Consumer Financial Protection Bureau, The Consumer Credit Card Market (2025 report to Congress)