Where You Buy Your Used Car Matters
Published
The finding
After controlling for vehicle type, age, and mileage, used car prices in Alaska run 10.8% above what you'd expect to pay for an equivalent vehicle nationally. In New Jersey, they run 5.7% below it. That is a 16.5-percentage-point spread between the most and least expensive markets in the country.
The gap is real. Of the 50 states in this dataset, 36 are distinguishable from the national average at 95% confidence — meaning the result is unlikely to be a statistical fluke. The remaining 14 are not statistically significant given their sample sizes, and are treated as indeterminate.
The six most expensive states — Alaska (+10.8%), Hawaii (+10.1%), Idaho (+6.4%), Oregon (+6.1%), Montana (+5.8%), and Washington (+5.4%) — are all western states. The six cheapest — New Jersey (−5.7%), Florida (−4.6%), Georgia (−3.6%), Virginia (−3.6%), Alabama (−3.4%), and Ohio (−3.3%) — cluster in the South Atlantic, mid-Atlantic, and Great Lakes corridors.
In dollar terms: a 2024 Honda CR-V — one of America's most popular compact SUVs — lists for an average of $33,500 nationally. The same car, same age, similar mileage, runs roughly $3,600 more in Alaska (around $37,100) and roughly $1,900 less in New Jersey (around $31,600). The spread between the two markets, for the same vehicle, is about $5,500.
Price premium by state
A state's price premium is how much its dealer asking prices run above or below the national average for similar vehicles — same type, same age, same mileage range.
66,750 listings across 50 states · April–May 2026
States ranked by price premium
66,750 listings · April–May 2026 · red = above average · blue = below average · whiskers = 95% CI
What the data shows
The dominant pattern is a western premium. All six of the most expensive states are in the American West: five — Alaska, Hawaii, Oregon, Montana, and Washington — on the Pacific Coast or the Canadian border, and Idaho within the Pacific Northwest interior. Maine (+4.5%) ranks seventh despite sitting on the Atlantic, though it too shares a land border with Canada. The clearest geographic exceptions in the top ten are Nebraska (+4.3%) and New Mexico (+4.2%), neither of which has an obvious claim to the western cluster. Why those states rank where they do is a question the data raises but cannot answer.
The cheapest markets run through the South Atlantic and mid-Atlantic. New Jersey (−5.7%) and Florida (−4.6%) anchor the bottom. Georgia (−3.6%), Virginia (−3.6%), and North Carolina (−1.6%) continue the pattern up the coast. New York (−1.4%) and Connecticut (−1.4%) are moderately below average. New England is divided: Maine ranks among the most expensive states in the country while neighboring New Hampshire (−1.8%) falls below the national average — two adjacent states pointing in opposite directions.
The two largest auto-producing states — Michigan (−3.0%) and Ohio (−3.3%) — both fall in the cheapest third of markets nationally. Indiana (−2.0%) is also below average. Pennsylvania (−0.7%) is modestly below. All four states most associated with domestic auto manufacturing sit on the inexpensive side of the distribution. Whether that reflects something structural about those markets is a question this dataset cannot resolve.
The Great Plains states lean expensive: Nebraska (+4.3%), South Dakota (+2.7%), Oklahoma (+2.2%), Kansas (+1.7%), and Texas (+2.6%) are all above the national average, as are Arkansas (+2.7%) and Mississippi (+2.0%) in the Deep South. The Mountain West is more mixed. New Mexico (+4.2%), Nevada (+3.8%), and Colorado (+1.2%) are above average. Wyoming — geographically western, rural, and remote — sits at +1.2% but is not statistically distinguishable from the national average. Utah, at −1.7%, falls on the opposite side. What separates these two states from the broader western pattern is unclear.
Alaska and Hawaii both warrant a specific note. Alaska leads the dataset at +10.8%, with a confidence interval of +8.8% to +12.8% across 521 listings from 18 dealers. Hawaii follows at +10.1%, CI +8.2% to +11.9%, across 656 listings from 30 dealers. For both states, the lower bound of the confidence interval clears zero by a larger margin than any other state in the dataset. If you are buying in either market, the data offers no version of this analysis in which you are not paying a premium.
Methodology & caveats
Price premium figures are derived from 66,750 dealer listings collected in April–May 2026, joined to dealer address records using high-confidence geocoding only. All 50 states are represented.
The core method compares each listing's price — expressed as a fraction of the vehicle's new-car sticker price (derived from NHTSA/EPA records) — to what similar vehicles sell for nationally: same type, same age range, same mileage range, within each vehicle class independently. The result is a state premium that cannot be explained away by inventory mix. A state that happens to carry more trucks does not artificially appear more expensive simply because trucks hold value differently from cars.
Confidence intervals are 95% normal approximation intervals: ±1.96 × (standard deviation / √n). In plain terms, a state is classified as statistically significant if its interval does not include zero — meaning the result is unlikely to be a statistical accident.
36 of 50 states meet the significance threshold. The 14 that do not — Rhode Island, Wyoming, Massachusetts, West Virginia, California, Minnesota, Wisconsin, Vermont, Pennsylvania, Kentucky, Illinois, Iowa, Delaware, and North Dakota — have overlapping confidence intervals that include zero. For these states, the sign of the estimate — positive or negative — should not be treated as a reliable finding.
A minimum of 200 listings per state was applied before analysis. Rhode Island (271 listings, 16 dealers) and Vermont (336 listings, 24 dealers) approach this threshold and carry wider uncertainty than larger states. No state was excluded by the minimum.
Dealer listings are not a random sample of all dealers. Whether dealers with structured digital inventory differ systematically from those without — in vehicle type, price point, or regional density — is unknown. If they do, the premiums reported here could be overstated or understated in ways this analysis cannot detect. Private-party listings (Craigslist, Facebook Marketplace) are not included. Data reflects inventory at the time of collection and will not be updated.