US Used Vehicle Market Report: September 2026

Market Reports · United States

By TradeBasis · Published September 8, 2026 · Data through August 2026

---

On the morning of September 8, 2026, two of the most widely cited wholesale benchmarks in the United States published their August numbers, and they pointed in opposite directions. Cox Automotive’s Manheim Used Vehicle Value Index came in at 208.2, up 0.4% against August 2025. Black Book’s Used Vehicle Retention Index came in at 144.3, down 1.5% against August 2025 and down for the third month in a row.

Neither index is wrong. This US used vehicle market report exists because the gap between them is the single most useful piece of information a dealer can have going into the fourth quarter. The two indices measure different slices of the age curve, and in August 2026 those slices moved in opposite directions for the first time in a meaningful way. The American used market has stopped behaving like one market. It is now at least three: a late-model band under pressure, an older affordable band that is outperforming, and an electrified band moving on its own clock.

What follows is the full August data set, the mechanics behind the divergence, and the specific appraisal and pricing adjustments it argues for in September. If you run a lot, wholesale for a living, or price retail inventory in a single market, the practical conclusion is short: a national average is now a worse guide to your next appraisal than it has been at any point in the post-pandemic period.

The August 2026 numbers, in one table

Every figure below is drawn from primary sources published between July 16 and September 8, 2026. Where a metric is reported on a lag, the reference month is stated.

Metric	Latest	vs. prior month	vs. year ago
Manheim Used Vehicle Value Index (Aug 2026)	208.2	-0.9%	+0.4%
Manheim non-adjusted wholesale prices (Aug 2026)	—	-0.5%	+0.2%
Black Book Used Vehicle Retention Index (Aug 2026)	144.3	-1.2%	-1.5%
MMR Three-Year-Old Index (Aug 2026)	—	-1.8% (typical: -0.6%)	—
MMR retention	99.5%	+0.3 pts	-0.3 pts
Auction sales conversion	57.2%	+3.5 pts	-2.8 pts vs. 3-yr Aug avg
Manheim Electric Vehicle Index	—	-4.7%	+4.5%
Manheim Non-EV Index	—	-0.6%	-0.9%
Wholesale days’ supply (end of Aug)	27 days	-1.4 days	+2.3 days
Retail used days’ supply (end of Aug)	44 days	-2 days	—
Average used listing price (July 2026)	$27,028	flat	+6%
New-vehicle ATP (July 2026)	$49,855	+0.2%	+1.9%
National average gasoline (Sept 4, 2026)	$4.15/gal	+1.6%	record for September
Rental risk unit values (Aug 2026)	—	-0.3%	+3.4%

Sources: Cox Automotive (Manheim Used Vehicle Value Index, August 2026, published September 8, 2026; used-vehicle inventory report, July 2026; Kelley Blue Book ATP, July 2026), Black Book (Used Vehicle Retention Index, August 2026), AAA (national and state fuel averages, September 4, 2026).

Why the two national indices disagree — and why that matters more than either number

Start with what each index actually measures, because the difference is the whole story.

The Manheim Used Vehicle Value Index is built from completed sales at Manheim’s US auctions across all 20 market classes in the J.D. Power classification scheme, drawing on more than five million transactions annually. It is adjusted for mix, mileage and seasonality. Critically, it spans the entire age curve, from nearly-new units through vehicles well past ten years old.

The Black Book Used Vehicle Retention Index is built from Black Book’s published wholesale average on 2- to 6-year-old vehicles only, expressed as a percentage of original typically-equipped MSRP, weighted by registration volume and adjusted for age, mileage, condition and seasonality.

Now overlay what Cox Automotive’s chief economist Jeremy Robb described in the August release. Depreciation, he noted, was not uniform: younger age cohorts fell more than usual, while older and more affordable units depreciated far less, with the 9-year-old and 10-year-old cohorts holding roughly six percentage points better than their long-term averages. The MMR Three-Year-Old Index dropped 1.8% in August against a typical August decline of 0.6%.

Put those two facts together and the divergence resolves cleanly. Black Book’s index sits entirely inside the age band that underperformed. Manheim’s index includes the older band that outperformed. One index went down 1.5% year over year; the other went up 0.4%. Both are accurate measurements of the market. They simply are not measurements of the same market.

This is not a statistical curiosity. It is a direct, quantifiable warning about appraisal error. If your book value, your trade allowance guide, or your buy-bid discipline is calibrated on late-model wholesale behavior — which most are, because that is where the volume and the data quality historically sat — then in September 2026 you are carrying a systematic bias in two directions at once:

• You are overpaying for 2- to 5-year-old trades. Three-year-old units shed value three times faster than a normal August. A book republished on a weekly or biweekly cycle is describing a price that no longer exists by the time the car is on your lot.
• You are underpaying for 8- to 12-year-old trades. The affordable end of the market is where demand has concentrated, and where retention is beating the long-term curve by roughly six points. These are the units your competitors are also undervaluing, which is precisely what makes them the opportunity.

The honest caveat: index divergence of this kind can narrow as quickly as it opened. September through November is historically the steepest seasonal depreciation stretch of the year, and if the older cohorts give back their outperformance the two indices will re-converge. That is an argument for measuring your own market weekly, not for picking whichever index flatters your inventory.

Demand did not weaken. It relocated.

It would be easy to read a softening wholesale month as a demand problem. The retail data says otherwise. Used retail sales picked up through August, and retail days’ supply fell to 44 days by month end, down two days from July. Dealers kept coming to the lanes to source product, which is why wholesale values firmed in the final weeks of the month after a weak start.

What changed is not how many Americans are buying used vehicles. It is which used vehicles they can afford. Three structural forces are pushing demand down the price ladder simultaneously:

1. The new-vehicle market is not providing an escape valve

The average new-vehicle transaction price was $49,855 in July 2026, up 1.9% year over year and just under the all-time peak of $50,609 set in December 2025. Prices have hovered near $50,000 for the better part of a year. Every month that new-vehicle affordability stays where it is, more demand is displaced into the used market — and specifically into the used market’s lower price bands.

2. The fleet is old, and old fleets generate forced purchases

The average age of a light vehicle on US roads reached a record 12.8 years in S&P Global Mobility’s most recent annual analysis, the eighth consecutive annual increase, with passenger cars averaging 14.5 years and light trucks 11.9 years. Vehicles in operation stand near 289 million with a steady 4.5% scrappage rate. As Cox noted in the August release, an aging parc means more owners facing large repair bills, and a meaningful share of them respond by replacing the vehicle rather than fixing it. That is demand that arrives on a repair-shop timeline rather than a discretionary one, and it lands overwhelmingly in the sub-$20,000 tier.

3. Fuel costs are now a first-order variable again

The national average price of regular gasoline was $4.15 per gallon on September 4, 2026 — a record for the month of September and up from roughly $3.14 a year earlier. Cox flagged fuel directly as a driver of segment performance in August. Compact cars were the strongest mainstream segment year over year, EV values held a 4.5% annual gain, and pickups and SUVs declined.

None of these three forces is cyclical noise. All three point the same way: toward smaller, older, cheaper, and more efficient units holding value better than the market average, and toward large, thirsty, late-model units carrying more depreciation risk than a national index implies.

Segment performance: the affordability rotation is now a year old

The August segment picture is the clearest expression yet of a rotation that has been building through 2026. Overall market prices were up slightly year over year, led by EVs and compact cars, while pickups and SUVs continued to trend down.

Two months earlier, in mid-August’s preliminary read, compact cars were the strongest performer at plus 2.2% year over year while the luxury segment had narrowed to plus 0.7% after leading for much of the first half. That reversal matters. Luxury led the market through the spring on the strength of electrified premium product. As the year progressed, buyers rotated toward value, and the leadership moved down-market.

EVs: strongest year-over-year gain, sharpest monthly drop

The used EV story in 2026 has been counterintuitive enough that it deserves its own paragraph. The federal clean vehicle credits — the $7,500 new-vehicle credit and the $4,000 used-vehicle credit under IRS Section 25E — expired on September 30, 2025. The widely predicted collapse in used EV values did not happen. Through the first half of 2026, used EV wholesale values rose, not fell. Two forces explain it: record fuel prices pulled cross-shoppers into electric, and used EVs stopped competing against new leases whose payments had been artificially suppressed by a credit the lessor could capture.

That trend is now moderating rather than reversing. The Manheim EV Index closed August 4.5% above a year earlier but 4.7% below July — two consecutive months of decline as off-lease EV supply builds. Off-lease EV returns were projected to rise sharply in 2026, with a larger wave expected in 2027.

For an appraiser, the practical fact is the volatility, not the direction. A segment that can move nearly five percentage points in a single month cannot be priced from a book that republishes monthly. On a $30,000 used EV, one month of index lag is roughly $1,400 — larger than the gross on most used deals.

Trucks and SUVs: the risk is concentrated here

Pickups and SUVs posted year-over-year declines in August while the Non-EV Index fell 0.9% against last year. This is the segment where the fuel-price environment, the affordability rotation, and the late-model depreciation problem all overlap on the same vehicle. A three-year-old full-size pickup in September 2026 sits in the fastest-depreciating age band, in a softening segment, in a record-high fuel environment. Any two of those would justify a more conservative buy bid. All three together justify pricing it against live comparable listings rather than any published guide.

The regional variable most national reports ignore: fuel price dispersion

National fuel averages conceal a spread wide enough to change which vehicles are worth owning. On September 4, 2026, California averaged $5.81 per gallon while Indiana averaged $3.45 — a spread of $2.36, or roughly 68%. Oregon posted the largest monthly increase in the country at 7.2%, reaching $5.00, while eight states recorded declines.

Run that through ownership cost. A driver covering 15,000 miles a year in a 17 mpg full-size pickup buys roughly 880 gallons annually. At Indiana prices that is about $3,040 a year. At California prices it is about $5,110. The same truck, the same mileage, a difference of more than $2,000 per year in running cost — before a single dollar of the purchase price is financed.

That gap does not stay in the fuel budget. It shows up in what a payment-constrained buyer will bid for the truck, which shows up in days-to-turn, which shows up in what the unit is worth to you at appraisal. A national segment index reports one number for a vehicle whose real-world cost of ownership varies by four figures annually depending on the state it is registered in.

The operational conclusion is direct: fuel-sensitive segments — full-size pickups, large SUVs, V8 performance, and on the other side, hybrids and EVs — should be appraised against local comparable listings, not national guide values. In a $4-plus fuel environment with a $2.36 state spread, geography is a pricing input, not a footnote.

Supply: tight at retail, loosening at wholesale, and about to loosen further

Retail and wholesale supply are moving in opposite directions, which is itself informative.

Retail days’ supply fell to 44 at the end of August, down two days, on the back of improving used retail sales. Wholesale days’ supply was 27 days, down 1.4 days from July but 2.3 days higher than a year ago. Dealers are selling through and restocking, which is why wholesale values firmed late in the month even as the monthly index printed lower.

The forward pressure comes from leasing. Cox Automotive’s weighted lease equity measure turned negative in August 2026, at roughly negative $66, having been positive as recently as the prior period. That number sounds trivial and is not. Lease equity is what determines whether a maturing lease gets bought out by the lessee or the dealer, or gets returned to the grounding lender and pushed into the wholesale lanes. When average equity is positive, a large share of maturities never reach auction. When it turns negative, they do.

Layer on the volume forecasts. Off-lease returns were projected to rise materially in 2026 across all powertrains, with EV lease returns forecast to surge in 2026 and a still larger wave expected in 2027. The combination of negative lease equity and rising maturity volume is the clearest identifiable source of additional wholesale supply between now and year end.

Black Book’s Laura Wehunt flagged the other fourth-quarter pressure directly: manufacturers historically increase incentives in Q4, and heavier new-vehicle incentives have historically pressured used values. September through November is already the steepest seasonal depreciation window of the year.

The consumer balance sheet is the ceiling on retail pricing

Every wholesale forecast eventually runs into the question of what the retail buyer can actually finance. In September 2026, that answer is constrained in ways worth stating precisely.

• 29.6% of trade-ins toward new-vehicle purchases carried negative equity in Q2 2026 — nearly three in ten — up from 26.6% in Q2 2025 and the highest second-quarter reading since 2020.
• The average amount owed on those underwater trades was $6,884, a record for a second quarter, following $7,183 in Q1 2026.
• Buyers rolling that debt forward paid an average of $944 per month in Q2, against an industry average of $777, and are projected to pay an average of $16,270 in interest over the life of the loan.
• Auto loan delinquency is elevated. New York Fed data showed 5.6% of outstanding auto debt at least 90 days delinquent in Q1 2026, above the 5.3% peak recorded in Q4 2010.
• Loan terms are stretching to absorb it. In Q1 2026, 90.2% of new loans involving a negative-equity trade carried terms of at least 72 months and 43% extended to 84 months.

Edmunds’ Ivan Drury made the point that matters most for appraisers: the largest dollar losses are showing up on trucks and sedans that have traditionally held value well, which makes this a financing problem rather than purely a residual-value problem.

For a dealer, this has two concrete consequences. First, a meaningful share of the trades walking onto your lot arrive with roughly $7,000 of debt attached, which compresses what any appraisal number can actually accomplish in the deal. Second, retail price ceilings are set by monthly payment, not by sticker. In a market where the payment is the binding constraint, the store that prices most accurately against its true local competitive set wins the deal — and the store that prices off a national guide gives away either gross or turn, usually without knowing which.

Policy watch: what is actually still in force

Trade and incentive policy has moved repeatedly since 2025, and a fair amount of secondhand summary in circulation is out of date. As of September 2026, the items that still bear on US used-vehicle values:

• Section 232 automotive tariffs remain in place. A 25% tariff applies to imported passenger vehicles, light trucks and a broad range of automotive parts, with preferential treatment generally limited to qualifying USMCA-originating content. An import adjustment offset for parts duties steps down from 3.75% of aggregate US-built MSRP to 2.5% for vehicles assembled from May 1, 2026.
• The IEEPA tariff line was disturbed and partly replaced. The Supreme Court held in February 2026 that IEEPA did not authorize those tariffs; many were subsequently re-implemented under temporary statutory authority, which expired in July 2026, after which new tariffs were implemented under different authority. Section 232 auto duties were not disturbed by that ruling.
• Federal EV purchase credits are gone. Both the new-vehicle credit and the Section 25E used-vehicle credit expired September 30, 2025, and no federal replacement exists. State and utility-level incentives still vary widely and are worth checking market by market.
• The USMCA review is on the 2026 calendar, and Section 232 automotive measures are expected to feature in it.

The used-market effect of tariffs has been indirect but persistent: by holding new-vehicle prices near $50,000, they keep displaced demand flowing into used inventory. Cox noted that tariff impacts had supported higher wholesale values in the prior year’s comparison period, which is part of why the August year-over-year comparison was the toughest of 2026.

What this used vehicle market report means for your September appraisals

Seven adjustments the August data specifically argues for. Each is tied to a number above.

1. Re-book the 2- to 6-year-old band on a weekly cycle, not a monthly one

Three-year-old MMR values fell 1.8% in one month against a 0.6% norm. On a $28,000 unit that is roughly $500 of value that disappears between one book refresh and the next. This is the band where stale data costs real money right now.

2. Bid more aggressively on the 8- to 12-year-old band

Nine- and ten-year-old cohorts are holding roughly six percentage points better than their long-term depreciation average. This is where demand concentrated, and where most competing appraisals are still applying a normal-market depreciation assumption. It is the clearest mispricing in the current market.

3. Price fuel-sensitive segments to your state, not to the nation

A $2.36 spread between the cheapest and most expensive state fuel averages translates into four-figure annual ownership-cost differences on a full-size truck. Trucks, large SUVs, hybrids and EVs need local comparable pricing.

4. Treat used EVs as a live-comparables problem

A 4.7% single-month move is larger than the gross on most used deals. Any EV valuation older than a week should be treated as an estimate, not a number.

5. Assume negative equity behind the trade until proven otherwise

With 29.6% of new-vehicle trade-ins underwater by an average of $6,884, the appraisal conversation is frequently a debt conversation. Knowing your number is defensible — and being able to show the customer the comparable listings behind it — shortens that conversation considerably.

6. Watch conversion, not just price

Auction conversion at 57.2% sits 2.8 points below the recent three-year August average. Roughly two in five units offered did not sell. If your acquisition plan assumes the lanes are a reliable exit for anything you misjudge, that assumption is weaker than it was last year.

7. Set retail price against your actual competitive set, and re-check it weekly

At 44 days’ retail supply and 57.2% wholesale conversion, both the cost of overpricing and the cost of underpricing are elevated. The only pricing input that reflects both your segment and your geography is what comparable vehicles are listed and selling for inside your own radius.

Outlook: what to expect through the fourth quarter

The base case for October through December is continued depreciation from a structurally elevated level, with the split-by-age pattern persisting or widening.

Supporting continued softness: the September-to-November seasonal window is the steepest of the year; Black Book has flagged rising Q4 manufacturer incentives as additional pressure and forecast 11.9% depreciation for 2026 overall; negative lease equity should push a larger share of maturities into wholesale; and off-lease volume is rising.

Supporting a floor under values: retail demand has been resilient and days’ supply is tightening, not building; the vehicle parc keeps aging and generating replacement demand; new-vehicle prices near $50,000 continue displacing buyers into used; late-model used supply remains structurally constrained by weak 2020-2022 new-vehicle production; and Black Book’s point stands that values remain roughly 30% above 2019 levels.

The realistic expectation is a normal seasonal fourth quarter in aggregate, with unusually wide dispersion underneath it. Black Book put this well in its refreshed forecast: the used market no longer behaves as a single system, but as a collection of distinct segments each responding differently to the same conditions. That is precisely the environment in which a national average is least useful and a local competitive set is most valuable.

Frequently asked questions

What are used car prices doing right now in September 2026?

Wholesale used-vehicle values are essentially flat year over year and softening month to month. The Manheim Used Vehicle Value Index finished August 2026 at 208.2, up 0.4% against August 2025 and down 0.9% from July. Black Book’s Used Vehicle Retention Index, which tracks only 2- to 6-year-old vehicles, fell to 144.3, down 1.5% year over year and down 1.2% from July, its third consecutive monthly decline. Retail listing prices remain high: the average US used-vehicle listing price was $27,028 in July 2026, roughly 6% above a year earlier.

Why do Manheim and Black Book disagree about used vehicle values?

Because they measure different parts of the market. Manheim’s index covers all 20 market classes across the full age curve, adjusted for mix, mileage and seasonality. Black Book’s Retention Index measures only 2- to 6-year-old vehicles as a percentage of original MSRP. In August 2026 the younger age bands depreciated faster than normal while 9- and 10-year-old vehicles held roughly six percentage points better than their long-term average. Black Book’s index sits entirely inside the underperforming band, so it fell. Manheim’s index includes the outperforming older units, so it did not. The gap between them measures how far the market has split by vehicle age.

Are used car prices going up or down for the rest of 2026?

Most indicators point to continued depreciation through the fourth quarter, from a structurally high base. September through November is historically the steepest seasonal decline of the year. Black Book expects depreciation to continue and has flagged the Q4 increase in manufacturer incentives as additional downward pressure on used values, with a full-year 2026 depreciation forecast of 11.9%. Values remain roughly 30% above 2019 levels, so a return to pre-pandemic pricing is not the expectation.

Which used vehicle segments are holding value best in 2026?

Affordability and fuel economy are driving the winners. In August 2026, compact cars were the strongest mainstream segment year over year and the Manheim Electric Vehicle Index was up 4.5% against August 2025. Pickups and SUVs posted year-over-year declines and the Non-EV Index was down 0.9%. Older, cheaper units are also outperforming, with 9- and 10-year-old cohorts depreciating roughly six percentage points less than their long-term averages.

Are used EV prices falling now that the federal tax credit is gone?

Not year over year, but the direction changed over the summer. The federal clean vehicle credits, including the used-EV credit under IRS Section 25E, expired September 30, 2025. Used EV wholesale values rose through the first half of 2026 anyway, helped by record fuel prices and by used EVs no longer competing against heavily subsidized new leases. In August 2026 the Manheim EV Index was still 4.5% above a year earlier but fell 4.7% in a single month as off-lease EV supply built. Used EV values now move faster than any book can be republished, which makes live comparable listings the only reliable pricing input for that segment.

What is the average used car listing price in the US right now?

The average US used-vehicle listing price was $27,028 in July 2026, up about 6% year over year and effectively unchanged from June, according to Cox Automotive’s analysis of vAuto Live Market View data. July was the second consecutive month above $27,000, the first back-to-back stretch above that level since the summer of 2023.

How many days’ supply of used vehicles do US dealers have?

Retail used-vehicle days’ supply fell to 44 days at the end of August 2026, down two days from the end of July, as used retail sales picked up through the month. Wholesale days’ supply was 27 days at the end of August, down 1.4 days from July but 2.3 days higher than August 2025.

How should dealers appraise trade-ins in the current used vehicle market?

Appraise by age band and by local market rather than off a single national number. Late-model units in the 2- to 6-year-old band are depreciating faster than the long-term average, so book values on those vehicles go stale on the way down. Older units in the 8- to 12-year-old band are retaining better than normal and are frequently undervalued. Fuel-sensitive segments should be priced against the local fuel environment, since state gasoline averages ranged from $3.45 to $5.81 per gallon in early September 2026. In every case the reliable input is the live competitive set: what comparable vehicles are actually listed and sold for inside your own market radius, refreshed weekly.

The Bottom Line

August 2026 produced a genuinely unusual result: the two most-cited wholesale benchmarks in the United States moved in opposite directions year over year. Manheim finished at 208.2, up 0.4%. Black Book finished at 144.3, down 1.5%. The reconciliation is that the market has split by age, and the two indices sample different halves of that split.

Underneath, the pattern is consistent and it has been building all year. Payment-constrained American buyers are rotating toward what they can afford: older, smaller, cheaper, and more fuel-efficient. Compact cars and EVs are outperforming. Pickups and SUVs are not. Nine- and ten-year-old units are holding six points better than their long-term curve while three-year-old units depreciate at three times the normal August rate. Record September fuel prices, a $2.36 spread between the cheapest and most expensive states, near-$50,000 new-vehicle transaction prices, a record-old vehicle fleet, and nearly three in ten trade-ins arriving underwater by an average of $6,884 are all pushing in the same direction.

For anyone putting a number on a vehicle in September 2026, the conclusion follows directly. The national average has never described your lot, but the cost of pretending it does has risen sharply. The defensible number is the one built from what comparable vehicles are actually listed and selling for in your own market, in your own segment, in your own state — and refreshed often enough to keep up with a market moving this unevenly.

Machine-readable data and related resources

This report is published alongside a structured dataset for analysts, AI systems and automated tools. It is free to use under CC-BY-4.0 with attribution to TradeBasis (tradebasis.net).

Resource	What it contains
TradeBasis US Market Index (JSON)	Monthly machine-readable dataset: national metrics, segment performance, EV values, supply, credit conditions, policy, outlook and dealer action items.
llms.txt	Index of TradeBasis US research for AI systems and retrieval tools.
TradeBasis US	Live vehicle appraisal, market valuation and competitive-set pricing for US dealers, independents and wholesalers.

Sources

• Cox Automotive, “Manheim Used Vehicle Value Index: August 2026 Trends,” September 8, 2026.
• Cox Automotive, “Manheim Used Vehicle Value Index: Mid-August 2026 Trends,” August 19, 2026.
• Cox Automotive, “Manheim Market Insights, August 2026” (Cox Automotive Lease Equity data), August 2026.
• Black Book, Used Vehicle Retention Index, August 2026, reported September 8, 2026.
• Black Book, refreshed 2026 depreciation and residual value forecast, May 2026.
• Cox Automotive, “Used-Vehicle Inventory Edges Higher in July, but Affordable Options Remain Scarce” (vAuto Live Market View), August 2026.
• Kelley Blue Book / Cox Automotive, new-vehicle average transaction price report, July 2026.
• AAA, national and state gasoline price averages, September 4, 2026.
• Edmunds, Q2 2026 negative equity report, July 16, 2026; Q1 2026 loan-term analysis, April 2026.
• Federal Reserve Bank of New York, Household Debt and Credit Report, Q1 2026 (auto loan delinquency).
• S&P Global Mobility, average age of US light vehicles, most recent annual analysis.
• Congressional Research Service, “Section 232 Automotive Tariffs: Issues for Congress,” 2026.
• U.S. Internal Revenue Service, Section 25E previously-owned clean vehicle credit (expired September 30, 2025).

---

TradeBasis is a used-vehicle appraisal and market intelligence platform for dealers, independents and wholesalers, operated by Basis Software Inc. TradeBasis publishes monthly US market analysis at insights.tradebasis.net. This report is provided for informational purposes and reflects data available as of September 8, 2026.