Scott Langlinais | April 1, 2010
The Five-Step Approach to Fraud Detection is a strategy I use to detect fraud in any area, and a template I provide to company executives and managers when helping them establish control systems designed to detect frauds in their day-to-day operations. This is the second in a series of articles in which I will demonstrate how you can apply this strategy to your own environment.
Here is the Five-Step Approach.
Step one halts most people because if you have no idea what can go wrong in your area, the rest of the strategy collapses. This continues a series of articles in which I will walk through some very common and dangerous frauds that affect all organizations, regardless of industry, to help you understand how to apply the strategy to create an environment hostile toward fraud.
By this we mean when executives and managers manipulate revenues around a period end to craft an earnings figure that is more in line with either stakeholder expectations or their compensation stipulations. Whether I am performing a tactical review of an area or discussing fraud-prevention strategy with executives, I always begin with a "What Can Go Wrong" list, in which I list potential perpetrators and fraud acts. Considering the risk of executives managing earnings relative to revenues, here are a few good examples of what can go wrong.
Typically, my What Can Go Wrong documents for a particular area will list at least two or three dozen frauds stated in a single sentence or two. It is important to list both the perpetrator and the fraud act when you create your own exposure lists. Resist the urge to eliminate the perpetrator; their inclusion in your list brings the fraud to life and gives your list a sense of action.
Earnings management frauds such as these are perpetrated by high-level folks, and can result in millions of dollars in fines by the Securities and Exchange Commission (SEC) and severe market damage. MicroStrategy's executives learned this in 2000 when the SEC brought civil charges against the company's executives for improper revenue recognition. The top three executives were ordered to pay a total of $11 million in fines and penalties to the SEC, while the company's stock price plummeted from $260 per share to $86 in a single day of trading and continued to decline thereafter. 1
Too often we focus on the easy targets—the clerk in the corner rather than the company's rainmakers. But your most dangerous frauds will be those perpetrated by executives, so be sure to include them as potential perpetrators.
The next step in the process is to list the symptoms, or what these frauds would look like in the books and records. Here is a short list derived from the frauds listed above—you are likely to come up with many more.
You will notice that I did not list a single control weakness. A control weakness is not a symptom of fraud. Just because a control is present does not mean a fraud is not occurring. Conversely, just because a control is absent does not mean a fraud is occurring. Likewise, just because someone smokes does not mean they have lung cancer, and just because they do not smoke does not mean their lungs are clear. A doctor must look for the symptoms, as should we.
This is the last step I will discuss in the five-step approach to fraud detection; the other two are self-explanatory. If you perform audits, your step here is to include symptom detection in your audit programs. Auditors: look for symptoms of fraud! Quit looking for approval signatures and thinking your work is done; every fraudulent disbursement or expense report I have seen in my career had an approval signature on it. This does not mean someone approved the frauds, it just means the approver failed to pay attention, did not take their authority seriously, did not have time to properly review the item, or did not understand (or care about) what they should have been looking for.
If you manage an operational or finance/accounting unit, then design processes to detect symptoms. Managers generally understand how to establish preventative controls: approval signatures for checks over a certain amount, requiring original receipts on expense reports, three-way matching approved purchase orders to invoices to packing slips. But managers are not so good at establishing processes to detect frauds after the perpetrator has run the gauntlet of front-end controls. It is like a rancher who builds a fence around his livestock but has no way to catch the thief who has jumped the barrier.
Following are some audit tests/detective processes designed to catch the symptoms listed above. In each of these cases, you will need to ensure proper documentation exists around the sale, including a contract dated in the same period as the revenue recognition. Also—this is an important procedure—confirm the sale and terms with the customer, by phone, not by boilerplate letter or email.
Of course, the descriptions of these tests are too general to properly implement, but they should provide you with an idea about how to construct detective procedures within your own environment. Good luck in discovering symptoms of earnings management!
Opinions expressed in Expert Commentary articles are those of the author and are not necessarily held by the author's employer or IRMI. Expert Commentary articles and other IRMI Online content do not purport to provide legal, accounting, or other professional advice or opinion. If such advice is needed, consult with your attorney, accountant, or other qualified adviser.
Footnotes