Buyer questionAnswered from the China side
Supplier financial risk in China: there are no statements to analyse.
The standard procedure — pull the accounts, compute liquidity and leverage ratios, compare to sector norms — assumes the supplier publishes accounts. A privately held Chinese manufacturer does not, and no amount of paying for data changes that. What exists instead is a set of public distress signals: enforcement actions, regulatory penalties, abnormality listings, equity freezes. They are genuinely useful and they are not a financial analysis. This page is about the difference, because treating the second as the first is how buyers end up feeling informed while knowing very little.
FINANCIAL RISK / 01
Absence of bad news is the trap in this whole exercise.
Every signal below is asymmetric: its presence is informative, its absence is close to meaningless. Read the section on what stays unanswerable before using any of it in a decision.
Why the standard playbook does not transfer
Financial risk frameworks — whether a credit report, a ratio model or a procurement questionnaire — are built on the assumption that audited accounts exist and can be obtained. For listed companies anywhere, including China, they can. For the privately held factories and trading companies that most buyers actually deal with, they cannot: there is no public filing obligation that would put a profit-and-loss statement or a balance sheet into anyone’s hands.
Two consequences follow, and both are routinely missed. First, any “financial risk score” offered for such a company is a model output built on proxies, not a reading of its finances — and the proxies are usually the same public signals described below. Second, a supplier questionnaire asking for financial statements will be answered with whatever the supplier chooses to prepare, which is not an audited document and cannot be checked against a filing.
Registered capital is the most misread field in this whole area. It is the amount shareholders have committed, not money in a bank account, and it can be committed on a schedule running years into the future. A large registered capital is not evidence of resources. What registered capital does and does not mean.
What is genuinely public
These come from state disclosure systems, and each resolves to a record with a date:
Business-abnormality listing (经营异常名录) — an administrative flag, most often for failing to file an annual report or for being unreachable at the registered address. Mundane in isolation, and worth asking about when the company is otherwise presenting itself as substantial.
Administrative penalties — regulator sanctions. The subject matter matters more than the count: an environmental penalty at a factory you are about to place a production order with is a different fact from a late-filing fine.
Court enforcement and judgment records — a company that is being enforced against has creditors who went to court and won. Where a company appears as a defendant repeatedly in payment disputes, that is the closest thing to a public solvency signal that exists.
Equity freezes and pledges — shareholding encumbered by court order or pledged as security. Both indicate that someone else has a claim on the ownership of the business.
Former registered names, address and shareholder changes — individually routine, collectively a picture of how stable the entity has been. How to read change records without over-reading them.
Reading the signals without over-reading them
Presence is informative; absence is not. A company with three enforcement records has told you something. A company with none has told you only that nothing has been published against it — which is also true of every company that is failing quietly, and of every company whose creditors have not yet sued.
Recency beats count. One enforcement record from last quarter is a stronger signal than four from six years ago.
Republication lags. Commercial platforms mirror official disclosures, and the mirror is not instantaneous. “No record on this platform at this retrieval date” is the honest formulation; “this company has no penalties” is not.
Size changes the meaning. A large manufacturer will accumulate routine litigation the way any large business does. The question is not whether records exist but whether they are about failing to pay people.
What stays unanswerable, and what we do not do
Being direct about this is the point of the page:
We do not analyse financial statements. For a non-listed Chinese supplier there are none to analyse, and we will not build a ratio analysis out of proxies and present it as one.
We do not issue a credit score or a rating. A single number would imply a model we do not have and a data set that does not exist.
We do not predict failure. Cash position, order-book concentration, whether their own customer just cancelled — none of it is public, and all of it matters more to whether your deposit is safe than anything that is.
What we do is narrower and checkable: retrieve the public records, report counts and dates with their source, and say plainly when something could not be read rather than returning a clean-looking blank.
What to do with the answer
Because the signals are asymmetric, they are better used to set terms than to make a go/no-go call:
Structure the payment. Deposit size and balance timing are the levers that actually limit exposure, and they cost nothing to negotiate. This matters more on a first order than any assessment does.
Confirm the beneficiary. A supplier under financial pressure is where payment-diversion requests appear — a “temporary” account, a third-party company, an individual. Check the beneficiary against the registered entity.
Do not skip pre-shipment inspection. Corner-cutting shows up in the goods before it shows up in a court record.
Re-check before the large order. These records change. The check that mattered was the one run before the money moved, not the one from last year.
Not legal, financial or investment advice. Records are reported with their source and retrieval date, and a record check is not a judgement on a supplier’s solvency.