EUIPO's €700 Renewal Price: When Abandoning a Dormant EUTM Pays

EUIPO cut the ten-year EU trade mark renewal fee from 850 to 700 — a 17.6% discount on keeping a mark alive. The discount flatters a habit, not an asset. Any mark left unused for five consecutive years stands exposed under Article 47(2), its blocking power already decayed; for those marks, the rational renewal price was 0 long before the fee schedule said 700.

Banking prices the same neglect more bluntly. Banks and card issuers commonly charge $5–$15 per month in inactivity fees, gift cards bleed around $2.50 per month, and one documented $50 gift card was whittled down to $5 before its owner noticed. Dormancy fees exist because issuers accept what trademark owners tend to resist: an unused asset keeps generating carrying costs whether or not it retains strategic value.

The Rover marque shows the holding side of that ledger: considered dormant since 2005, it has changed hands repeatedly rather than lapsed — a standing bill for a name no factory uses. An estimated $58 billion meanwhile sits unclaimed in dormant American accounts, down from $80 billion, evidence that unwatched value eventually gets swept. The 700 fee simply lowers the rent on all of it. The rational default flips to audit-then-prune: pay only where the mark still blocks a live competitor.

The 700 Price List: What a Renewal Buys

The first line of every EUIPO renewal invoice reads 700 for the first class, with further per-class charges for each additional class — down from 850, 50, and 150 under the previous schedule. According to EUIPO's fee schedule, that payment buys a full 10-year term for every class on the register, whether or not the mark has ever been used. Most owners read the reduction as a flat discount. Run the arithmetic across portfolio width and it inverts: the saving shrinks with every class added, reaches zero at four classes, and turns negative at five.

A five-class renewal now costs 50 more than the identical filing did in 2024. The new schedule rewards narrow marks and quietly taxes broad ones — precisely the broad registrations most likely to be carrying dead weight.

The deadline architecture gives you twelve months of runway to make that call. The renewal request can be lodged up to 6 months before expiry at the standard fee; miss it and a 6-month grace period opens at a late-payment surcharge — enough to turn 700 into 875 on a one-class mark. A mark left unrenewed through the grace period is deleted from the register the day after it closes, with no further payment option. Every notice landing on a desk prices out at the new schedule.

Here is the structural quirk that drives the economics: EUIPO conducts no use audit at renewal. Nobody asks for invoices, sales channels, or a single shipped product; the payment re-certifies classes the owner already knows are empty. Because pricing is per class, each dead class carries its own line item — but that same per-class pricing supplies the fix. Article 51 EUTMR permits partial surrender, so dead classes can be stripped before paying. A three-class mark with one live class renews with the two dead shells billed alongside it; surrender the shells first and the invoice drops to 700.

What the receipt does not buy is time on the use clock. Under Article 47(2) EUTMR, a mark unused for 5 consecutive years anywhere in the EU is vulnerable to revocation — and the renewal payment does not reset that clock. It runs from registration or the last genuine use, not from the renewal date. Paying therefore does not preserve enforceable exclusivity: for a mark dormant past the 5-year wall, the freshly issued certificate is a revocable shell that any competitor can pierce with a routine cancellation action.

The same provision hollows out defensive value. When a mark older than 5 years is cited in opposition, the applicant can demand proof of genuine use, and failure to produce it kills the opposition outright. A dormant renewed mark still blocks new filings procedurally — it forces a response — but it excludes nothing substantively.

That is the lens worth adopting: treat each EUTM as an asset with a 700-per-decade carrying cost and a blocking value that decays toward zero at the 5-year wall. Read that way, the renewal notice stops being an administrative chore and becomes a valuation exercise — EUIPO is asking you to repurchase, for another decade, exclusivity you may no longer hold. Whether the rational bid is the full fee, a trimmed one, or zero is the exit question the sections that follow resolve.

Deadwood on the Register

When Ganglmair, Helmers and Love mined multi-decade UK trademark filings for the International Review of Law and Economics, roughly half the marks in their sample showed no detectable use at all. Treat that share as the baseline for how much of any mature register sits dormant. It reframes what a renewal notice actually is: not a bill for an asset, but an invoice asking whether you want to keep paying rent on inventory that may have no tenant — and with the renewal fee now trimmed, light rent on an empty unit is still rent.

The scale at EU level comes from EUIPO's own reporting: its annual report tracks the yearly inflow of new EUTM applications. That inflow continuously feeds the dormant pool from which renewal notices are generated. The mechanics mirror the dormant savings account that consumer guidance warns about — an abandoned account isn't closed, it's dormant, and it can still generate charges. EUIPO's renewal engine bills dormant marks on schedule regardless of whether anything commercial is happening behind the registration number.

Two court rulings complicate any aggressive prune-the-register instinct. First, the floor: in Leno Merken, the CJEU held that genuine use in a single Member State can suffice for the entire EU. A mark moving trivial volume in one Benelux country is legally alive across every Member State — invisible on a global use dashboard, yet fully defensible in a cancellation fight. Second, the opposite tolerance: in Vodafone v Orange, the General Court accepted invoicing-level, defensive use as genuine, sustaining a near-dormant mark whose function was blocking competitors. Courts will prop up registrations whose purpose is exclusion, which means dormancy is a spectrum, not a binary — and adjudicators sit at the forgiving end of it.

So the audit sequence matters. Before classifying a mark as deadwood, pull per-Member-State sales traces and invoice files; either artifact can resurrect a registration you were about to surrender. If neither exists — zero genuine use anywhere in the EU for five-plus years — the renewal payment buys nothing enforceable. Kill the comfortable assumption here: paying to renew does not preserve exclusivity. Under Article 47(2) EUTMR, the freshly stamped certificate is a revocable shell that any competitor can pierce with a routine non-use cancellation, and its opposition value collapses the moment an adversary demands proof of genuine use.

The EU trade mark reform sharpened the abandonment arithmetic. By removing the general requirement to prove use of an earlier mark in relative-ground invalidity actions — bad faith excepted — the reform stripped most of the shield value a dormant registration once offered, and lowered the legal cost of abandoning a mark today and cleanly re-filing tomorrow if strategy changes. Hoarding dead classes as insurance got cheaper to stop doing.

Apply the guide's standing rule: abandon only on confirmed five-plus-year EU-wide non-use with no launch scheduled within 24 months; in every other case, surrender the dead classes and renew the survivors. Sort each mark into a posture before the renewal window opens:

Four Exits from Dormancy

A dormant EUTM stands behind four doors, and the register prices them in different currencies: 900 a decade to keep every class, 700 a decade to keep one, a low-four-figure credit to hand the whole file to a stranger, or zero to walk away. Two of those doors end in the same room. Which exit dominates comes down to two variables only: the age of your last genuine use, and whether a relaunch is actually scheduled.

The instruments are few and their clocks unforgiving. Renewal runs through Article 49 EUTMR: the window opens six months before expiry, stays open through a post-expiry grace period (a surcharge attaches to late payment), and a mark left unpaid is deleted the day after that grace period closes. Partial surrender under Article 51 EUTMR is the only exit available at any point in the cycle — you strike individual dead classes from your own registration whenever the class-level autopsy says they're finished. An assignment takes effect when it is recorded against the Register, which makes the sale clock purely transactional. And lapse is the null option: file nothing, pay nothing, and EUIPO performs the deletion for you.

Cost separates the doors sharply. Keeping both classes of a two-class mark alive costs 900 per decade; surrendering the dead class and renewing the survivor runs the standard first-class rate of 700 per decade. A sale inverts the sign of the ledger: dormant word marks placed with brokers typically clear in the low four figures, because buyers price dormancy signals rather than documented use — the same signal-based valuation that governs dormant digital handles, where marketplace guidance warns that true abandonment is nearly impossible to prove. Lapse costs nothing in fees; the only optional spend is vigilance — a paid commercial EU watch service at CompuMark or Corsearch list pricing — or 0 if you sweep TMview yourself.

Blocking power is where the 900 invoice stops making sense. A live mark under five years old blocks hard: challengers cannot reach for the non-use weapon yet, so the registration deters filings on faith alone. Cross the five-year wall and the same registration blocks weakly — any applicant or canceller can demand proof of genuine use under Article 47(2) EUTMR, and a dormant mark has none to give. A lapsed mark blocks nothing at all. Read those three lines together and the convergence is stark: past the wall, a freshly renewed but unused mark and a deleted mark deliver essentially identical defense. The fee no longer buys exclusivity; it buys occupancy.

What renewal still buys, uniquely, is friction. Delete the mark and anyone can file the exact sign tomorrow morning — an EUTM application costs 700 and demands no use at filing, so the barrier to squatting your identity equals the fee you declined to pay. Keep it registered and a squatter must instead route around you: design away, seek consent, or spend months unwinding your shell in cancellation. For a sign with residual pull, that speed bump carries real option value; for a truly dead sign, nobody is queuing to steal it. Selling splits the difference: the dormancy problem transfers intact to a buyer who has already priced it into the offer.

Score the profiles and the winners fall out cleanly. The dormant profile — zero genuine use anywhere in the EU for five-plus years, no launch scheduled within 24 months — belongs to lapse: 0 versus 700+ per decade, for blocking power the wall has already flattened to near-nil. Where the mark is under five years old, or some classes still earn their keep, trim-and-renew wins. And whenever a willing buyer exists in the mark's Nice class, sell wins outright.

One caveat keeps the ledger honest: lapse paired with a paid watch service can cost more than renewal over a decade — so lapse dominates only in its native 0, manual-TMview configuration. The public register trail shows even blue-chip holders reaching this fork: BMW owned the Rover mark from 2000 to 2006 while licensee MG Rover supplied the actual use until 2005; when the user collapsed, the registration became a disposal decision, not a renewal line item. Before your next Article 49 window opens, pull the TMview use record, test broker appetite in your Nice class, and only then choose which door closes behind you.

What the Data Doesn't Tell You

Every abandonment decision inherits a measurement error, and the error runs in one direction: registry-based deadwood studies can observe public web presence and filed specimens, but they cannot see a Belgian distributor's purchase orders, a private-label supply contract, or spare-parts shipments that keep a mark technically alive. According to the Ganglmair, Helmers and Love dataset published in the International Review of Law and Economics, "use" was operationalized as *detectable* use — which makes their figures a floor, not a census, and the floor was measured on UK filings, not on the EUTM register where your renewal notice now sits. Treat any headline dormancy rate as a lower bound on uncertainty, never as a base rate for your own portfolio.

The second limitation is censoring. A mark renewed because a relaunch is quietly funded looks identical, in every dataset, to a mark renewed out of inertia — right up until the product ships or doesn't. Registry data records decisions, never intentions, so it cannot separate the rational renewal from the irrational one. That separation is precisely what the decision rule demands, and no observational dataset supplies it.

Variance across cases is driven almost entirely by challenger incentives, not by anything printed on the certificate. Two legally identical shells behave in opposite ways: one parked in a class crowded with litigious rivals gets pierced at the first opposition where the adversary demands proof of genuine use; the other, sitting in a niche class nobody contests, survives untouched for a decade. Practical fragility varies even where legal fragility is total — so the expected cost of holding a shell is a function of who else holds filings near yours, not of the fee schedule.

Three documented breaks in the rule, each traceable to named authority:

The working takeaway: the decision rule is only as sound as the use audit beneath it. Before signing either way, run the two-document check — a member-state revenue split and the last bona fide delivery record. The published data tells you the average case; your invoices tell you yours, and the rule breaks exactly where the two diverge.

What the Deadwood Data Hides

The deadwood headline above was not computed on EUTMs at all. Ganglmair, Helmers and Love built their ratio from UKIPO filing records — a register thick with sole traders, local partnerships, and single-shop brands for whom national filing was the cheap, obvious move. The EUIPO register self-selects differently: an applicant who paid a premium for simultaneous coverage of every Member State has, by that very act, filtered for actual or planned cross-border distribution. Importing the UK ratio onto the EU register is an extrapolation the underlying data cannot confirm — useful for intuition, worthless as an EU-specific base rate.

The measurement error also cuts the other way, and that direction is expensive. Where the previous section mapped what registries cannot observe, the legal consequence is sharper: web-based use detection reads websites and marketplaces, but not the printed B2B catalogue an industrial supplier mails quarterly, the export invoices for goods shipped wholly outside the EU, or the intra-group transfers that European case law credits as the proprietor's own use. Under the Court of Justice's Leno Merken ruling, genuine use in a single Member State sustains the entire EUTM. Some certified deadwood is therefore legally alive, and pruning it destroys value no dashboard can see.

The variable that actually dominates the decision appears in no registry field: the product roadmap. Renewal records show payments, never intentions. The arithmetic is unforgiving — against a relaunch failure cost that runs to six figures once rebranding, scrapped packaging tooling, and a lost seasonal window are tallied, a 700 renewal breaks even at a launch probability well under one percent. No public dataset reveals that probability; only the roadmap memo does. This is precisely why the decision rule above demands a scheduled launch, not a hunch, before the renewal branch opens.

Sector averages misprice almost by construction. Pharmaceutical and technology filers deliberately hold defensive blocker marks through regulatory timelines no renewal calendar controls — a molecule awaiting authorization, a platform awaiting certification — so their dormancy is strategy, not neglect. Fashion and FMCG marks live or die by shelf presence, where dormancy signals genuine death. A register-wide dormancy average lands between two populations for whom it is close to meaningless, and allocating an abandonment budget off that average errs in opposite directions for each.

Abandonment also hands the exact sign to whoever wants it. A squatter files at the mirror-image 700 application fee, receives a fresh five-year grace period from the new registration date, and gains standing to oppose the original owner's future filings, sue its relaunch, and lodge customs actions. Buy-backs settle at multiples of that application fee: Apple spent years unwinding Shenzhen Proview's "iPad" registrations in China, and Anheuser-Busch's Budweiser fight with Budějovický Budvar has outlasted generations of counsel on both sides. When both conditions of the rule are met, that tail is a priced cost of exit — dominance means least-cost exit, not costless exit — and it is a reason borderline marks belong in the renew-the-surviving-classes branch.

Finally, the deterrence blind spot. A freshly renewed certificate is a revocable shell — Article 47(2) EUTMR lets any competitor pierce it after the statutory non-use window — but shells still read as walls to unsophisticated infringers who never challenge validity. Marketplace infrastructure amplifies this: Amazon's Brand Registry, for one, admits only holders of an active registered mark, and takedown programs run on registration status, not proof of genuine use. Deadwood statistics count challenges actually filed; they record none of the listings never posted because a certificate hung on the wall. Quiet deterrence is real, unmeasured, and incapable of saving a verified shell.

Across all six blind spots, one instrument outperforms every public dataset: the holder's own file. Before this cycle's renewal deadline, assemble three artifacts — Member-State-level use evidence including catalogues, export paperwork, and intra-group transfers; a signed roadmap statement covering the next two planning cycles; and a squatting-exposure check on the exact word and figurative elements. Run the decision rule above on those artifacts, not on a dormancy dashboard. The register tells you what a renewal costs; only your own records tell you whether anything is left to buy.

Also worth reading: 7 Critical Updates in USPTO's New Beta Trademark Search System for 2025: 7 Critical Updates in USPTO's · USPTO Unveils Enhanced Trademark Identifier Search Tool for 2025 Applications: USPTO Unveils Enhanced Trademark Identifier · Trademark Search Automation in 2025 7 Key AI Tools Revolutionizing USPTO Database Analysis: Trademark Search Automation in 2025