Finding Hidden Risks Before the Market Does

Hidden risks are almost never hidden in the sense of being secret. They sit in public filings, in the parts of the document that are dense, dull, and easy to skip: the footnotes, the off-balance-sheet commitments, the concentration of revenue in a few customers, the contingent liabilities parked in the notes, and the quiet creep in working capital. The market reacts to these later, when one of them turns into a headline. The evidence was usually there on the day the filing came out.

So finding risk early is less about cleverness and more about method. You go looking in the same handful of places every time, you read the parts most people scroll past, and you write down what you find before you decide what it means. The focus here is where to hunt and how to do it repeatably. It is the companion to why forensic analysis matters, which makes the broader case for not taking reported numbers at face value.

Why risk hides in plain sight

A set of financial statements is built to summarise. The income statement gives you one revenue line and one profit line. That compression is useful, but it is also where exposure disappears. A company can earn most of its revenue from two customers and still show a single healthy topline. It can carry a large disputed tax demand and still report clean profit, because the demand lives in the notes, not on the face of the accounts.

The result is that the riskiest facts are often the least prominent. They are disclosed, because disclosure rules require it, but they are disclosed in the places designed for completeness rather than attention. Reading only the headline numbers is like reading only the chapter titles of a book. You get the shape of the story and miss the part where it turns.

Most risks that later become headlines were footnotes first. The work is reading the footnotes.

The five places to hunt

There is no single detector for hidden risk, but there is a short, reusable list of where it tends to accumulate. Work through these in order for any company you take seriously.

1. The notes and footnotes. The notes to the accounts are where the real detail lives: how revenue is recognised, what a segment actually contains, how a large receivable is being treated, what changed in an accounting policy this year. A change in policy that shifts more revenue into the current period, or a quiet reclassification, will show up here long before it shows up in a ratio. This is close cousin to reading between the lines of annual reports, which goes deeper on the subtext of disclosure language.

2. Off-balance-sheet items. Some obligations do not appear as debt on the balance sheet but still commit the company to future cash: operating lease commitments, guarantees given on behalf of subsidiaries or joint ventures, letters of comfort, and capital commitments already contracted. A business can look modestly geared on the face of it and carry substantial off-balance-sheet obligations in the notes. The question to ask is simple: what has this company promised to pay that is not sitting in the debt line?

3. Customer and supplier concentration. If a large share of revenue comes from a handful of customers, the loss of one changes the whole picture. The same is true on the supply side: a single supplier for a critical input, or a single plant, is a fragility. Concentration is not automatically bad, and many strong businesses are concentrated by nature. The point is to know the exposure and watch it, rather than be surprised by it when a contract is not renewed.

4. Contingent liabilities. These are possible obligations that have not yet crystallised: disputed tax demands, pending litigation, claims, and guarantees. They are disclosed in the notes precisely because their outcome is uncertain. Most never convert. But their size relative to annual profit is worth knowing, because the ones that do convert can be large, and they were visible in advance. Reading the contingent-liability note every year, and watching whether the total is drifting up, is basic hygiene.

5. Working-capital creep. This is the slow one, and the easiest to miss because it happens across periods rather than in a single line. When receivables or inventory grow faster than sales for several quarters, cash gets tied up even while reported profit rises. Sometimes the reason is structural and ordinary. Sometimes it is a warning that sales are being booked that will not turn into cash, or that inventory is not moving. The creep itself is the signal to investigate. The mechanics of how receivables, inventory, and payables move through a business are covered in working capital and the cash conversion cycle.

Working capital as an early warning: one worked signal

Working-capital creep is worth a closer look because it is where a hidden risk and a visible number meet. Consider a jewellery business over recent financial years, using its public accounts. In one year its operating cash flow was roughly half its reported profit. In the next it was actually negative even though profit was positive, meaning the business reported a profit but consumed cash from operations. The year after, operating cash flow recovered to slightly above profit.

The explanation is not fraud and not a scandal. A jewellery business ties up large amounts of cash in gold and store inventory as it grows, so cash and profit diverge sharply from year to year. That is a structural feature of the business, not a defect. But notice what the working-capital lens did: it turned a smooth-looking profit line into a question, “where did the cash go this year,” and the honest answer was inventory.

That is the whole discipline in miniature. The number flagged something. You went and found the reason. The reason here was benign and explainable. In another company the same signal might lead to a less comfortable answer, and you would have found it early. Note that this is the signal, not a verdict on the company. The deeper method of comparing cash to profit belongs to stock forensics; here we are only using the creep as a trailhead for risk.

A repeatable hunt, not a one-time check

The reason to make this a routine rather than a one-off is that risk is not static. A contingent liability that was small three years ago can grow. Customer concentration can rise as one client scales faster than the rest. Off-balance-sheet commitments accumulate. The value of the method comes from doing it every reporting period and comparing to the last, so you are watching the direction, not just the level.

A simple table helps keep the hunt consistent from one company to the next.

Where to look	What you are checking	The question it answers
Notes and footnotes	Policy changes, revenue recognition, reclassifications	Did the accounting quietly shift this year?
Off-balance-sheet	Leases, guarantees, capital commitments	What is owed that is not in the debt line?
Concentration	Revenue by customer, critical suppliers or plants	What single loss would change the picture?
Contingent liabilities	Disputed tax, litigation, claims	What could crystallise, and how big?
Working capital	Receivable and inventory days versus sales	Is cash getting tied up faster than sales grow?

None of this requires special access. It requires reading the parts of the filing that are already public and that most readers skip. One caution when you hold this year’s disclosure against an old one: prior-period comparables get restated after demergers, accounting changes, or segment redefinitions, so make sure you are comparing like with like.

The mindset that makes it work

The hardest part of this is not the technique. It is the willingness to keep reading past the point where the story already looks fine. A clean headline invites you to stop. The discipline is to treat the clean headline as the beginning of the check rather than the end of it, and to write down what you find so that next period you can see what moved.

Most of what you turn up will be ordinary and explainable, and that is the point. You are not trying to find a scandal in every company. You are trying to make sure that when something does start to bend, you see the footnote before it becomes the headline.