Editor’s Note — DCC.
This is the foundational piece Wang Qinglan wrote in September 2024 arguing that data registration cannot confirm rights, because the law has not yet conferred data rights. It is the conceptual predecessor to her December 2025 piece on the SPC’s new ‘data disputes’ case category — the SPC’s procedural move forced the doctrinal question into the open, but Wang had identified the gap a year earlier. For overseas counsel approaching the Chinese data property rights regime, this is the doctrinal essay; the SPC piece is the operational sequel. DCC’s framing emphasizes the civil-law doctrine that overseas teams trained in common-law systems will find counterintuitive.
”Confirming rights” — what does the verb actually do?
In Chinese legal usage, 确权 (literally, “confirming rights”) means confirming the attribution and nature of a right. The verb assumes a right exists; confirmation is a downstream act on that pre-existing right.
This is the conceptual frame that breaks when applied to data.
Wang’s central observation: there is no legal right in data to confirm. The Chinese legal community has not converged on what kind of right “data” is — property right? intellectual property? a new species of property altogether? The NPC has not legislated a defined right. Without legislation, no legal data right exists.
The state’s response, per Wang: “Carry on arguing — we won’t wait for you. But the data-element market can’t wait either. So we’ll set aside the rights debate and use the property rights (产权) framework to protect data property interests and promote data circulation.”
The Data 20 Articles, December 2022, established the structural-subdivision data property rights regime:
- Data resource holding right (数据资源持有权)
- Data processing-and-use right (数据加工使用权)
- Data product operation right (数据产品经营权)
The three rights are the C-stage stars of the Chinese data economy. Each is registrable. Each gets a certificate.
But each is a property right (产权), not a legal right (权利).
The civil-law doctrine that makes the distinction matter
In Chinese civil-law doctrine, only the NPC and its Standing Committee can confer civil rights (赋权). The Data 20 Articles is a Central Committee + State Council policy directive — outside the NPC’s legislative authority. It can establish a property rights regime (产权制度) — but property rights are an economic concept, not a civil-law concept. The same conceptual move appears in the Third Plenum of the 20th CCP Central Committee Decision (July 2024), which speaks of “accelerating the establishment of data property attribution determination” — not “rights determination.”
“Accelerate the establishment of data property attribution determination (数据产权归属认定), market transaction, benefit allocation (权益分配), and interest protection systems…”
— Third Plenum Decision, 2024
Wang’s emphasis: the policy text uses attribution determination (认定), not rights determination (确认). The vocabulary is precise. The Chinese drafting team knew this would be parsed against civil-law doctrine.
The two ways civil rights actually get confirmed
In Chinese civil-law practice, rights confirmation runs on two tracks:
Track 1 — “Friendly” confirmation
Rights confirmation that doesn’t require dispute. Two sub-types:
- Possession-based confirmation (占有确权) — for general movables. Whoever possesses the asset is presumed to hold the right. Every transfer of possession is also a transfer of rights. Wang’s example: you buy a used computer on Xianyu (Chinese eBay-equivalent). Whoever physically possesses the laptop at the moment is presumed owner — until possession transfers at the meeting point. The doctrine matches a real intuition: bought from someone, possession transfers, ownership transfers, the rights confirmation is automatic.
- Registration-based confirmation (登记确权) — for high-value assets where possession alone doesn’t provide enough public notice. The classic case is real estate: buying a house requires the formal registration step at the registry office, and ownership transfers only when registration completes. The mortgage on a house must also be registered to be effective.
Track 2 — “Adversarial” confirmation
When parties dispute attribution, the court confirms the right through litigation. The right is named and assigned to the prevailing party. This is the adversarial mode and works for all rights types — property rights, contract rights, IP rights — once the rights themselves are statutorily defined.
The full doctrinal cycle: legislature confers → registry / court confirms → disputes resolve. The verb chain has no broken link.
Where data registration breaks the chain
The data property rights regime breaks the chain at the first link. The NPC has not conferred a data right. So:
- Possession-based confirmation of a data right doesn’t work — there’s no right to confirm.
- Registration-based confirmation of a data right doesn’t work either — registration “confirms” what the legislature created, and the legislature hasn’t created anything yet.
- Adversarial confirmation in court doesn’t work — the court can only adjudicate rights the legislature defined. As of Wang’s writing in September 2024, no Chinese court had cited the Data 20 Articles’ three property rights to confirm data rights in adjudication. Judges, operating under civil-law doctrine, cannot create rights through interpretation.
In Wang’s words: “In China — a civil-law country — judges can’t make law. They can only adjudicate under the law.”
What data registration is actually doing
Wang’s reframing: data registration is publicity (公示), not confirmation (确权).
The chain runs: trade → needs publicity → publicity creates public credit → public credit creates market trust → market trust enables trade. Registration is one form of publicity. The function is signaling to potential downstream counterparties that this asset’s chain of title is documented. The publicity supports market confidence — not legal rights creation.
Wang’s metaphor: registration is a trust credential (可信凭证), not a rights certificate (确权证书).
The operational difference:
- A rights certificate would have constitutive effect — without registration, no right exists. (Real estate ownership has this character in China.)
- A trust credential has evidentiary effect — it’s preliminary evidence that the registrant has documented its data chain-of-title and submitted to compliance review by the registration institution. It can be overcome by contrary evidence.
The 2024 Beijing IP Court case that made the doctrine concrete
In a case Wang flags — Datatang v. Yinmu (数据堂 v. 隐木, the “first case on the effectiveness of data IP registration certificates”) — the Beijing IP Court explicitly declined to recognize a Data IP Registration Certificate as an absolute property-right confirmation. The court confirmed only the certificate’s effect as a trust credential.
The court’s reasoning (Wang paraphrases): “Before a property-natured legal interest has been confirmed as an absolute property right by law, the holder of the property-natured interest cannot seek judicial protection by analogizing to other absolute property right types.” The phrase “by analogy” is the key. The court would not extend property-rights doctrine to a not-yet-confirmed-by-law category.
The trust credential effect held — but only as preliminary evidence (初步证据), defeasible by contrary evidence. This is meaningfully weaker than what most market participants understood the certificate to provide.
The three judicial protection paths that work today
Without a statutory data right, Chinese courts protect data interests through three existing-law analogies:
- Compilation work copyright (汇编作品保护) — if the data set demonstrates creativity in selection or arrangement, treat it as a compilation work under copyright law.
- Trade secret protection (商业秘密保护) — if the data set satisfies the trade-secret elements (secret, kept secret, has commercial value), protect under the Anti-Unfair Competition Law.
- Competitive interest protection (竞争性权益保护) — under the AUCL general clause, protect data as a competitive interest the data handler has invested in.
The three paths are workable but uneven. Compilation copyright requires creativity in selection. Trade secret protection requires secrecy. Competitive interest protection is the catch-all, but it’s a general-clause claim with substantial evidentiary burdens (and now operates under the 2025 AUCL Article 13(3) data clause). Wang’s prediction at the time: “data property rights” will not become full legal rights soon — but the trust-credential function of registration can still meaningfully support these three paths.
What this means for the registration regime going forward
Wang’s prescription, in September 2024:
- Registration institutions should stop overclaiming “confirmation.” Calling registration confirmation (确权) misleads the market about what a certificate does.
- Strengthen compliance review. A registration institution’s substantive review of data legality and authenticity is what gives the certificate evidentiary weight in court. Without rigorous review, the certificate has no evidentiary value.
- Use the attribution determination vocabulary. The Third Plenum’s attribution determination (认定) language is doctrinally precise — registration institutions can determine the attribution of data property without claiming to confirm legal rights.
- Build a regulatory framework. Trust-credential value depends on the credibility of the issuing institution. The state should regulate registration institutions to protect the market’s confidence in registration certificates as a category.
The SPC’s January 2026 “data disputes” case category change — which Wang followed up on — vindicated the underlying doctrinal critique. Once the SPC named the case category, courts gained a procedural channel to scrutinize data registration certificates directly. Wang’s September 2024 piece had warned exactly this gap was coming.
Why this matters for overseas teams
Three implications for foreign counsel advising on Chinese data deals:
- Don’t translate 确权登记 as “rights confirmation registration.” A more accurate rendering — “property attribution determination registration” — preserves the doctrinal distinction. Translating it as “rights” creates false expectations on both sides of the transaction.
- A registration certificate is one input to evidentiary strategy, not a title document. When advising on a Chinese data acquisition, the registration certificate is useful — it shows the seller invested in chain-of-title documentation. But it does not vest legal title in the buyer in any way comparable to a deed or a patent assignment. The buyer’s protection in a downstream dispute will run through compilation work, trade secret, or competitive-interest doctrine.
- Choose the registration institution carefully. Per the NDA’s draft Data Property Rights Registration Work Guide, only registration institutions practicing substantive review (not formal review) will produce certificates with meaningful evidentiary weight. The Shenzhen Data Exchange’s “dual-verification” model is the operational benchmark.
The deeper observation in Wang’s piece is that the Chinese data property rights regime is doctrinally incomplete by design — the state chose to operationalize a property rights system without waiting for the NPC to confer formal legal rights. Overseas counsel who expect the regime to behave like a Western IP system will be repeatedly surprised. The regime behaves like a publicity and credit-building system. Once that’s internalized, the operational logic falls into place.
— Wang Qinglan (王青兰), 数据确权登记，谁给的勇气？ (Data Rights Confirmation Registration — Who Gave You the Courage?), 青兰数据观察 WeChat Official Account, September 19, 2024. Original article (Chinese).
Not legal advice. The above is DCC’s structured summary of Wang’s commentary; not a verbatim translation. The author’s views are her own and do not represent her employer.