What This Index Is, and Why It Exists
Every quarter, our dealer feed accumulates tens of thousands of Hermès listings. Most of what we publish from that data is configuration-specific — our Birkin price guide and Kelly 25 price guide exist because most buyers only care about the price band for the one bag they want. This report does something different: it aggregates the whole feed into one comparable index across the two most recent quarters, so instead of “what should I expect to pay for a Birkin 25 right now,” the question becomes “is the resale market, as a whole, actually moving.”
We built it because trend claims about Hermès resale circulate constantly — “prices are exploding,” “the market is cooling” — almost always without a stated sample size, a stated time window, or a stated definition of “price.” This index states all three, is built entirely from our own dealer feed, and is intended to run as a recurring series: a Q3 2026 update is already in progress. That recurrence is the point — one quarter tells you a number, several quarters in a row tell you whether it means anything.
We are publishing it as a standalone, linkable reference rather than folding the figures into another guide. If you are a journalist, a fellow reseller or a researcher who wants a citable number for “how is Hermès resale actually moving right now,” that is exactly what this page is for — the chart, the underlying medians and the full method live on one page with a permanent URL and a citation line at the bottom, so a number never floats free of the sample and caveats behind it.
The Headline Finding: Flat, With a Soft Edge
Across the 7 model-and-size cells with a large enough sample to read with confidence, the average absolute quarter-over-quarter move was only 2.6% — but those moves didn't cancel out evenly. Net across the 7 cells, asking prices slipped 1.6%. That is still essentially flat — a 1.6-point net drift on a base this size is not a correction — but it is a flat market with a slight downward tilt, not a perfectly static one, and one cell accounts for most of that tilt: Birkin 35, down -9.8% to a median of €12,996 on 250→446 listings. The rest of the large cells were far quieter: Birkin 25 rose, Kelly 20 and Kelly 25 barely moved, Kelly 28 was flat, and Birkin 30 and Kelly Pochette eased only slightly.
That distinction — essentially flat, but with a soft edge rather than a dead flat line — is worth stating precisely because it cuts against both narratives that dominate secondhand commentary: the market is neither overheating nor cracking, but it is also not perfectly inert. For context on the retail side of that story, see our piece on how Hermès has historically raised boutique prices — the two trends (steadily rising retail, softly drifting resale over a single half-year) are not in conflict; they simply operate on different timescales.
The Index, Interactively
Every bar below is live data, not an illustration. Filter to the 7 large-sample cells or see all 16 tracked configurations in non-exotic leathers, sort by sample size, price or change, and hover or keyboard-focus any row for the full breakdown — including how spread out asking prices are within a quarter, not just where the median sits. Exotic skins are priced separately, further down this page, so they never distort the medians shown here.
Hover or focus a row to see full detail — median asking price, how many listings it's based on, and the spread of prices within that quarter. Showing 7 of 16 tracked configurations.
Each row compares one bag's median asking price across quarters: the hollow outlined bar is last quarter (Q1 2026), the solid filled bar is this quarter (Q2 2026).
View this chart as a table
| Model / size | Sample | Listings (Q1 → Q2) | Q1 median | Q2 median | Change | Middle half of listings (Q2) |
|---|---|---|---|---|---|---|
| Birkin 25 | Large | 1,161 → 2,125 | €20,800 | €21,500 | +3.4% | €19,350 – €23,500 |
| Birkin 30 | Large | 933 → 1,632 | €18,340 | €17,844 | -2.7% | €13,603 – €21,500 |
| Kelly 25 | Large | 791 → 1,552 | €20,000 | €19,885 | -0.6% | €18,000 – €21,537 |
| Kelly 28 | Large | 483 → 929 | €15,985 | €16,000 | +0.1% | €12,674 – €18,694 |
| Kelly 20 | Large | 313 → 537 | €22,935 | €22,810 | -0.5% | €21,000 – €25,218 |
| Kelly Pochette | Large | 244 → 564 | €18,500 | €18,263 | -1.3% | €16,564 – €19,650 |
| Birkin 35 | Large | 250 → 446 | €14,400 | €12,996 | -9.8% | €9,396 – €20,983 |
By Model: The Large-Sample Cells
These 7 cells each carry the largest samples in both Q1 and Q2 — large enough that the changes below are worth reading as directional signal, not noise.
Birkin 25 moved from a median of €20,800 to €21,500 (+3.4%, 1,161→2,125 listings) — the clearest riser in the index, and consistent with what we've written about the 25's outsized demand relative to boutique supply in our Birkin investment guide. Birkin 30 softened marginally, €18,340 to €17,844 (-2.7%). Birkin 35 is the outlier of the group — down -9.8%, from €14,400 to €12,996 on 250→446 listings. That is a meaningfully larger move than anything else in this section, and it is consistent with a market where buyer preference is tilting toward the more wearable 25 and 30 over the larger 35 — the 35's listing count nearly doubled quarter over quarter, which points to sellers responding to softer demand at that size rather than a data artifact.
Kelly 25 was flat in every sense that matters, €20,000 to €19,885 (-0.6%), and Kelly 20 did not move much — €22,935 to €22,810 (-0.5%). Kelly 28 was essentially unchanged (+0.1%, to €16,000), while the Kelly Pochette eased -1.3% to a median of €18,263.
Note the ranges in the chart above are wide relative to the median moves — for Birkin 25, the middle half of Q2 asking prices ran €19,350 to €23,500. Condition, leather and hardware configuration explain most of that spread, not the market moving under any single listing — our Kelly 25 price guide breaks down exactly what drives a bag to either end of that band.
Smaller Samples, Read as Indicative
The remaining 9 cells in the index carry smaller samples than the large-sample group above. We still report them — a smaller sample is still information — but we mark them small-sample and treat the changes as directional at most, never as a confirmed trend.
A few small-sample cells moved enough, on samples large enough to take seriously as early signal, to call out specifically: Kelly Mini rose +5.8% to a median of €22,548 (104→318 listings); Constance 18 eased -2.2% to €11,000 (145→246 listings); and Kelly Cut rose +7.7% to €19,000 (94→203 listings). Kelly 32 is the one worth watching most closely: it fell -15.9% to €8,730 (127→206 listings) — both the largest sample of any small-sample cell and now the largest move anywhere in the index, large or small. It doesn't yet clear our large-sample bar, so we are not calling it a trend, but it is the clearest early signal in this report worth confirming next quarter.
The remaining small-sample cells moved less dramatically this quarter: Kelly Danse (+5.0%), Constance 24 (-11.9%), Picotin 18 (+3.2%), Birkin 29 (+0.3%) and Birkin 40 (-1.8%). Constance 24's move is the largest among them, but on a sample this size a shift in which specific bags dealers happened to list — newer versus older, pristine versus well-loved — can look like a price move without being one. See Methodology below for why we raised our reporting floor to guard against exactly that, and why none of the cells above should be read as a confirmed trend on their own.
The Exotic Skin Premium
This report's strongest new finding isn't a quarter-over-quarter move at all — it's what happens when you stop blending exotic skins (alligator, crocodile, ostrich, niloticus, porosus and lizard) into the same median as standard leathers. Across all 7 model-and-size pairs where we have enough exotic-skin listings to report a median — 102–456 exotic-skin listings each — an exotic runs at 1.93x to 2.48x the standard-leather price this quarter, median 2.16x. That range is remarkably consistent — this isn't one model with an exotic-skin cult following skewing the average, it's a structural premium that holds across Birkin, Kelly and Kelly Pochette alike.
That structural, roughly 2x-plus premium is also exactly why blending exotic and standard leathers into one median — which is what most casual market commentary does, and what an earlier version of this report did too — produces misleading numbers. See Methodology below for how much that distortion was actually worth on Birkin 35, the cell it hit hardest.
Hover or focus a row to see the standard-leather and exotic-skin medians behind that multiple, and how many exotic listings it's based on.
Each row compares the same bag in two leather types this quarter: the muted dot is the standard-leather median, the rose-gold dot is the exotic-skin median, and the number is the exotic price as a multiple of the standard price.
View this chart as a table
| Model / size | Standard median | Exotic median | Premium | Exotic listings |
|---|---|---|---|---|
| Birkin 25 | €21,500 | €53,322 | 2.48x | 456 |
| Kelly Pochette | €18,263 | €43,747 | 2.40x | 268 |
| Kelly 25 | €19,885 | €42,999 | 2.16x | 216 |
| Birkin 35 | €12,996 | €28,083 | 2.16x | 111 |
| Kelly 28 | €16,000 | €34,072 | 2.13x | 102 |
| Kelly 20 | €22,810 | €46,000 | 2.02x | 153 |
| Birkin 30 | €17,844 | €34,393 | 1.93x | 352 |
A Note on the Collector Tier: Birkin 20 “Faubourg”
One configuration is deliberately excluded from every chart on this page: the Birkin 20, sold exclusively as the “Faubourg,” a boutique-rare micro-bag. On 130 listings this quarter, its median asking price is €196,068 — roughly 7.7 times the highest median of any bag in the index above. That isn't a data error and it isn't this quarter's biggest mover; it's a different product category, priced and traded more like a collectible than a handbag, and folding it into a “bag price index” would misrepresent both the index and the Faubourg itself. We track it separately and may cover it in its own report.
Methodology
We are publishing this in full because a data report that hides its method is not a data report — it's a claim.
- These are asking prices, not transactions. Every figure comes from live dealer listing text. We do not have visibility into final negotiated sale prices, and asking prices can and do differ from what a bag actually sells for.
- Source and deduplication. Data comes from Luxury Provider's own dealer feed. The raw feed contained meaningfully more rows than the 19,962 listings analysed here — we removed 14.8% of rows as reposts, identified by text-fingerprint deduplication (near-identical listing text reposted by the same or a different dealer), before running any of the statistics above.
- Currency handling. Prices were parsed from listing text across multiple currencies (EUR, HKD, GBP, USD, SGD, IDR) and converted to EUR using a single reference exchange rate applied identically to both quarters. That choice is deliberate: it means the quarter-over-quarter changes above reflect movement in sellers' local asking prices, not currency fluctuation between Q1 and Q2.
- Outliers and cell thresholds. Outliers were trimmed using the middle half of listings for that cell and quarter — the interquartile range, in statistical terms — so a single mispriced or joke listing far outside that band can't swing a whole cell's median. Only model-and-size cells with at least 40 listings in both quarters are reported at all; cells with smaller samples are marked as such rather than treated the same as the largest cells, and where exactly to draw that large/small line is a judgment call we chose to state rather than hide — a different threshold would move a handful of cells between the two groups, though it would not change the headline finding.
- Why we raised the reporting floor. This update raises the minimum sample size to 40 listings per quarter, up from a looser floor in an earlier draft. We made the change after tracing an outlier: a cell that no longer appears in this report — Kelly 35 — showed a +73.5% quarter-over-quarter move. We checked whether that was a parse error. It wasn't. Q1's sample for that configuration happened to be dominated by genuinely cheap, decades-old vintage pieces (P- and O-stamp, 1990s-era, visibly used), while Q2 skewed toward newer stock. The median moved because the age and condition mix of what dealers happened to list changed, not because prices did. That is a general risk on small cells, not a one-off: with fewer listings, a handful of unusually old or unusually pristine pieces can swing a quarter's median on their own, independent of any actual price movement. Raising the floor to 40 listings removed that cell and the other flakiest ones — the largest absolute move left anywhere in the 16-cell core set is now Kelly 32 at -15.9%, a fraction of the move we traced above. It is also exactly why only the 7 large-sample cells carry this report's headline number: that mix-shift risk shrinks, but never fully disappears, as sample size grows — small cells are still worth reporting, but never worth treating as confirmed.
- Exotic skins are split out, not blended in. Earlier drafts of this report combined exotic-skin listings (alligator, crocodile, ostrich, niloticus, porosus and lizard) into the same median as standard leathers for each model and size. Because exotics carry a consistent, large premium — see The Exotic Skin Premium above — that blending inflated the combined medians unevenly across cells, by as much as 25.6% on Birkin 35, the cell with the largest exotic-skin share relative to its total listings. We now report standard and exotic leathers as two separate figures for every model and size, which is why this update's numbers are not directly comparable to an earlier, blended draft of this index. Leather type was parsed from listing text; roughly 22% of listings didn't name a specific leather and are treated as standard, on the reasoning that an exotic skin is a strong selling point sellers almost always advertise explicitly — an unlabelled listing is far more likely to be an unremarkable standard leather than an unmentioned crocodile.
- Parse accuracy. We manually reviewed 50 randomly sampled listings against the prices our parser extracted to validate accuracy before trusting the pipeline at scale.
- This is a series, not a one-off. A Q3 2026 update is already planned. We think a single quarter is a data point; several in a row, published with the same method, is what makes an index trustworthy.
How to Cite This
You're welcome to reference the figures in this report — that's why we published the methodology in full. Please cite it as: Luxury Provider, “Hermès Resale Price Index: H1 2026,” accessed via https://luxury-provider.com/blog/market/hermes-resale-price-index-h1-2026. Linking directly to this page keeps the figures attached to the methodology and confidence labels behind them, rather than a number floating free of its context.
One last honesty note: this index measures what dealers are asking, in one channel, over one half-year. It does not measure auction results, private peer-to-peer sales, or boutique retail — and a single half-year of stability is not a promise about what Q3 or Q4 will show. We'd rather publish a smaller, well-labelled claim every quarter than a bigger one we can't stand behind.
See Today's Listings, Not Just the Index
The index shows where the aggregate market sits. Our collection lists individually authenticated Birkin, Kelly and Constance pieces with current asking prices stated transparently, configuration by configuration.