Carbuki Insights

Auto Credit Just Hit a 10-Year High. Here Is Why Your Store May Not Feel It Yet

July 13, 2026

Share of auto loans with terms longer than 72 months
June 202527.0%May 202630.0%June 202631.1%

June 2026 set a new all-time high in the dataset, up 110 bps from May and 410 bps from a year earlier. The May figure is implied by the reported monthly change. Source: Cox Automotive, Dealertrack Credit Availability Index, June 2026 (published July 10, 2026).

The number that moved

On July 10, Cox Automotive published its monthly read on auto lending, and the headline was the kind dealers have been waiting several years to hear. The Dealertrack Credit Availability Index rose to 104.6 in June 2026 - its highest level since December 2015, a fifth straight monthly increase, and up 7.3% from June 2025.

Underneath the index, the single largest driver was approvals. The overall loan approval rate climbed to 73.8%, up 170 basis points from May. That was the largest monthly gain of 2026 and the biggest single contributor to the index's rise.

For a dealer principal or GM, the temptation is to read that as a pure tailwind and move on. It is a tailwind. But the same report contains a second story, and it matters more for how a store is staffed and how its phones get answered: the yes is arriving with more strings attached than at any point in the history of the dataset.

Myth: When credit loosens, deals close themselves.

Data: In the same month approvals jumped to 73.8%, the share of loans stretching beyond 72 months hit an all-time high of 31.1%, down payments slipped to 13.2% (below the 13.7% of a year ago), and negative equity remained 220 basis points above year-ago levels.

  • Source: Cox Automotive, Dealertrack Credit Availability Index, June 2026 (published July 10, 2026)

What the June data actually says

The index tracks six factors. Reading them together tells a more useful story than the headline number does.

FactorJune 2026Change vs. MayChange vs. June 2025
Loan approval rate73.8%+170 bps+150 bps
Loans with terms over 72 months31.1% (all-time high)+110 bps+410 bps
Down payment share13.2%-30 bpsBelow 13.7% year-ago level
Subprime share of loans16.6%-10 bps+250 bps
Average contract rate10.98%+11 bps-
Yield spread6.77%+5 bps-43 bps

Source: Cox Automotive, Dealertrack Credit Availability Index, June 2026. The negative equity share eased 30 bps from 57% but remains 220 bps above year-ago levels, following March's record high of 59.2%.

Three of those lines are the reason a GM should not simply celebrate and move on.

Terms are stretching. Nearly a third of loans now run past 72 months - a record in the Cox dataset, and up 410 basis points from 27% a year ago. Long-term paper is what lenders and buyers reach for when the payment does not otherwise work.

Down payments are thinner. At 13.2%, buyers are putting less cash in than they did a year ago. Less money down means the payment has to carry more of the deal.

Negative equity is still elevated. Even after three consecutive monthly declines from March's record, the share of loans where the borrower starts underwater remains well above last year. Cox is blunt about what that means: most borrowers begin a new loan owing more than the vehicle is worth.

And borrowing costs did not improve. The average contract rate rose 11 basis points to 10.98%.

Approvals are a top-of-funnel event, not a delivery

The Credit Availability Index measures how easy it is to get a loan. It does not measure how many people buy. Those are different things, and the gap between them is where a dealership actually lives.

Looser underwriting expands the pool of shoppers who can be structured into a deal. It does not do the structuring. And on the evidence above, the structuring is getting harder, not easier. A deal built on a 75-month term, 13% down, and a rolled-in negative equity balance is not a deal that closes in one conversation. It closes in three or four: an initial payment quote, a re-quote after the trade payoff comes back, a lender counteroffer, a call asking what it would look like with two thousand down, and a confirmation.

That is the part nobody puts in a press release. Easier credit does not reduce the work per deal. It increases it - and it increases the number of shoppers generating that work.

The workload lands on the phone

More approvable shoppers multiplied by more complicated structures equals more touches per delivery. Those touches are overwhelmingly phone and text: payment questions, payoff questions, whether the approval came back, re-quotes, and appointment confirmations.

This is the constraint most stores hit first, and it is not a lender problem or a marketing problem. It is a capacity problem. A store can win the approval and still lose the customer because the callback took two days, or because the fourth inbound call at 5:40pm on a Saturday rang out. We have covered the arithmetic of that leak before, in the cost of missed calls at a dealership and in speed-to-lead in automotive.

The demand side supports the point. In the same week, the Cox Automotive Manheim report showed wholesale sales conversion at 57.5% in June - still 2.6 percentage points above the three-year norm for the month - with affordability described as the throughline of the first half of 2026. Shoppers are active and price-sensitive. Price-sensitive shoppers ask more questions. More questions mean more calls.

Where AI phone agents fit, and where they do not

It is worth being precise here, because the category is oversold.

An AI voice agent should not be quoting payments, making credit representations, or telling a customer what they qualify for. Those are human, compliance-bound conversations, and a store that automates them is buying a problem. Anyone evaluating AI calling should start with the regulatory picture, which we walked through in TCPA and AI calling compliance for dealerships.

What the technology does reliably is narrower and, in this market, more valuable:

  • Answer every inbound call, including the overflow, the after-hours calls, and the ones that arrive while the BDC is on the other line.
  • Capture intent and context - what the caller wants, which vehicle, what their trade is - so the human callback starts informed rather than cold.
  • Book the appointment when that is all the caller needs.
  • Route anything involving credit, payments, or payoff to a person, immediately and with the context attached.
  • Handle the mechanical follow-ups - confirmations, reminders, document reminders - that consume BDC hours without requiring judgment.

In a month where lenders said yes more often than they have in a decade, the marginal cost of losing a qualified, approvable shopper to a missed call goes up. That is the entire argument. It does not require believing that AI closes cars.

The counter-case, honestly stated

Two cautions belong in any sober read of this data.

First, the trend can reverse. The index has now risen five straight months, and the mix driving it - record-long terms, elevated negative equity, thinner down payments - is precisely the combination that has preceded underwriting pullbacks before. Cox flags long-term loan share and negative equity as ongoing watchpoints, noting that writing long paper to borrowers already underwater compounds duration and collateral risk. A store that builds its plan around permanently generous credit is building on sand.

Second, none of this is evenly distributed. Credit access improved unevenly by channel: independent used posted the largest monthly gain at 2.2%, while non-captive new actually declined 0.5%. Whether June's loosening shows up in your store depends on your mix and your lender relationships.

The useful conclusion is not that credit is easy, so sell more cars. It is that the binding constraint in a looser-credit month is rarely the lender. It is whether your store can absorb the resulting volume of conversations without dropping any - and that constraint is worth fixing in a tightening market too, which is rather the point.

What to watch next

  • The next index reading, published around the tenth of each month. Whether the approval-rate gain holds is the tell.
  • The share of loans beyond 72 months. Another record would suggest lenders are stretching rather than genuinely loosening.
  • Subprime share. It has now declined three months running. If it keeps falling while the index rises, the loosening is concentrated among better-qualified buyers - a narrower opportunity than the headline implies.

If your phones are the thing standing between an approvable shopper and a delivery, that is a solvable problem. Carbuki builds AI voice agents for US dealerships that answer every call, capture the context, and hand the credit conversations to your people.

Sources

Carbuki builds AI voice agents for retail automotive — answering sales and service calls, following up on leads, and booking appointments 24/7 in multiple languages.

See how it works →
Share:XLinkedInFacebookRedditEmail

← All articles