Top Analytic Tests Tutorial Series January 2014

What’s not to like about business meals? They’re delicious. And taking a client out for a night out on the town can make the difference between winning or losing a multi-million dollar contract, right?

So maybe it seems like we’re just being jealous when we’re flagging these exceptions, tirelessly serving our organizations from a windowless basement room next to the furnace.

The problem is, legislation like the Sunshine Act (for pharmaceutical and medical device companies) requires all attendees at business meals be recorded and reported. Other legislation, like the Foreign Corrupt Practices Act (FCPA) or the UK Bribery Act, is broadly applied and does not require such disclosure, but lavish and extravagant expenses can nonetheless be in violation. The potential impact is enormous: billions of dollars worth in fines have been handed out.

Large organizations must become well controlled around the recording of its business meal attendees. One place to start is monitor for business group meals where the average amount per attendee is greater than a specified threshold (say $75).

Risk: Documentation of group meal attendees is incomplete, creating a compliance or policy issue.

Test: Identify average amount of group meals per attendee; report cases where the average amount per attendee is greater than a specified threshold.

Data Acquisition:

Please refer to the previous article in this series on acquiring T&E expense data. It’s important to note that in order to do this test, you’ll have to have an application system for your expenses where the attendees are recorded.

We’ll assume that we can acquire the Expense Category (we’re going to only be interested in Group Meals), Expense Amount in Functional Currency (if not, we’ll need to do a conversion to your organization’s functional currency), and finally the Number of Attendees (if you have a list of attendees instead, you’ll need to prepare the data by first summarizing on the expense line to get the number of attendees).

Test Steps:

1. Obtain a dump of all the T&E from within the audit period you want to review.
   • See above section on Data Acquisition.
2. Import the expense data into ACL™ Analytics using the import wizard.
3. Ensure completeness by verifying the data against control totals (e.g., count, and total on expense amount to tie to audit period actuals).
4. Filter to include only group meals for our analysis.
5. Calculate the average amount per attendee by defining a computed field:
   • If the number of attendees is 0, then use Expense Amount (FC)
   • Otherwise, use Expense Amount (FC) / Number of Attendees
6. Filter and report any expenses where the average amount per attendee is greater than a specified threshold.

Conclusion:

This is a quick and easy way to start identifying potential omissions of attendees—these expenses will have an unusually high amount per listed attendee. You shouldn’t stop here, however. What if the user put duplicate attendee names for the same expense? What if only a first name was provided? Is there an opportunity to run the attendee list against the employee master to identify external attendees? There is a world of possibilities for attendee testing.

Effective superheroes have someone at HQ to help keep their gear in top shape, so if you’d like advice and/or guidance on implementing your own testing, you can have ACL’s experienced consulting team be your Lucius Fox to provide best practices, analytic design, and script development.

(Source: AuditNet)