Lesson 1 of 8
Five families
Manipulation has a small number of shapes. Sorting a suspicious figure into one of five tells you which disclosure to open next, and what it would show.
By Javier Sanz
Founder of ValueScreener. Built Ninety Nine, a securities services provider, and sold it to a tier-one global bank; ran operations and internationalization at Alpaca, the brokerage API.
Who writes thisEditorial processLinkedIn (opens in a new tab)X (opens in a new tab)
In one minute
- Dechow, Ge, Larson and Sloan examined 2,190 SEC enforcement releases issued between 1982 and 2005 and identified 677 firms alleged to have misstated their statements.
- In misstatement years, accruals and cash and credit sales ran unusually high while earnings and employee numbers were falling, and the firms were raising new financing.
- Sorting a figure into one of five families is useful because each family points at a different disclosure, and the disclosure either supports the figure or does not.
A suspicious number is not yet a finding. It becomes one when you can say what shape of manipulation would produce it, which disclosure would carry the evidence, and what that disclosure actually says.
Five families, and the disclosure each one points at
| Family | What is being moved | Where the evidence sits |
|---|---|---|
| Revenue that has not arrived | Timing of a sale, or its existence | The revenue note, shipping terms, and the receivables balance |
| Costs out of their period | Timing of an expense, or its classification as an asset | Depreciation lives, capitalisation policy, accrual and contingency notes |
| Cash flow relocated | Which of three sections a cash movement sits in | The cash flow statement's own classification and the supplemental disclosures |
| A measure the company defined | What is included in a figure with no accounting standard behind it | The required reconciliation to the nearest GAAP measure |
| A comparison a transaction reset | What the prior year is being compared against | The business combination note and its pro forma table |
What the largest study of the record found
Dechow, Ge, Larson and Sloan examined 2,190 Accounting and Auditing Enforcement Releases issued by the Commission between 1982 and 2005 and drew from them 677 firms alleged to have misstated quarterly or annual statements. They then compared those firms with themselves in other years, and with the broader population of listed firms.
Four of their findings are worth carrying into every lesson that follows. In misstatement years, accruals ran unusually high. Cash and credit sales ran unusually high. Earnings and the number of employees were declining. And the firms financed more of their assets through operating leases.
The motive they found, which is not the one people expect
The picture is not a failing company hiding a collapse. It is a company under market pressure: raising new financing, carrying a higher market-to-book ratio, and coming off strong prior share price performance.
That matters for where you look. A record of exceeding expectations, quarter after quarter, alongside a need to keep raising money, is the setting the enforcement record describes most often. Two of the cases in this course turn on exactly that pressure.
Why families rather than a list of tricks
A list of named tricks is unusable, because the next one has a new name. A family is durable because it is defined by what is being moved rather than by the mechanism.
There are only so many things a company can move. It can move revenue in time. It can move a cost in time, or out of the income statement entirely by calling it an asset. It can move cash between the operating, investing and financing sections without changing the total. It can define a measure that no standard governs. Or it can change what this year is being compared against.
The order of operations
Notice what the table does not contain: a conclusion. The suspicious figure sends you to a document, and the document either explains the figure or leaves it unexplained. An unexplained figure is a reason to keep reading, and the last lesson of this course puts a number on how often it turns out to be nothing.
Check yourself
In Dechow, Ge, Larson and Sloan's sample, what were misstating firms doing in the market during misstatement years?
- Trading at depressed multiples after poor share price performance
- Raising new financing, with higher market-to-book ratios and strong prior share price performance
- Delisting, in most cases
Show the answer
Answer: B.
The paper reports market pressure as the setting: new financing, higher market-to-book, and strong prior price performance, alongside declining earnings and employee numbers.Why organise the subject by family rather than by named trick?
- Because the names are hard to remember
- Because a family is defined by what is being moved, so it survives the invention of a new mechanism
- Because the SEC uses these five categories
Show the answer
Answer: B.
Mechanisms change and get new names. The set of things a company can move, in time or between sections or between definitions, is small and stable.Revenue grows strongly while operating cash flow falls. What has the reader established?
- That revenue was recognised too early
- Nothing yet: several families produce that pattern, and each points at a different disclosure
- That the company is a going concern risk
Show the answer
Answer: B.
The gap is a question, not an answer. Which document settles it depends on which family the cause belongs to.Sources
- Dechow, Ge, Larson and Sloan, "Predicting Material Accounting Misstatements": 2,190 SEC Accounting and Auditing Enforcement Releases issued between 1982 and 2005, and the 677 firms drawn from them.
- Fridson and Alvarez, Financial Statement Analysis: A Practitioner's Guide: statements are prepared by parties with an interest in the impression the statements create.
What this leaves out
- Every company named in this course is one that settled a public enforcement action or filed a public restatement, and every statement about it is confined to what that document says. Nothing in this course says or implies anything about any other company.
- The Dechow findings describe a sample of firms subject to enforcement between 1982 and 2005. They are associations in that sample, not tests any individual company passes or fails.
- The worked example uses invented round percentages and is labelled illustrative.
- Nothing here is advice about any security.