But then came Michelle Holt, a secretary with U.S. Air Force, Air Combat Command, Communication Support Squadron, at Joint Base Langley-Eustis. She took creative timekeeping fraud to the next level, raising the bar for the next fraudster.
Ms. Holt was a salaried employee on the General Schedule (GS) grade for the federal civilian workforce. As such, she was entitled to overtime pay if authorized by her employer, was also entitled to other forms of holiday and annual leave, and premium pay for any federal holidays worked.
According to this Department of Justice press release—
Holt falsely claimed over 42,000 hours in unauthorized overtime for hours she did not work, as well as other amounts of unauthorized holiday leave, sick leave and annual leave, all amounting to losses to the United States of more than $1.4 million. In recent years, Holt’s overtime pay was over double that of her regular salary.
The timekeeping fraud was perpetrated for nearly 17 years, apparently with nobody being the wiser that one employee had found a means to effectively triple her secretary’s salary through claiming overtime hours (among other things) for which she did not work. To put that fraud into perspective, that’s about 2,500 hours of overtime claimed per year. And nobody noticed.
How did Ms. Holt accomplish this feat? According to the DOJ—
Holt accomplished the fraud by making manual retroactive adjustments to protected computer time and attendance systems to add overtime, reverse leave taken and reverse holiday leave. In doing so, Holt used another employee’s log-in information without that employee’s knowledge or authorization.
Essentially, then, Ms. Holt falsified her identity with respect to accessing the timekeeping system and made “manual retroactive adjustments” to change her timekeeping records after they had been reviewed and approved by her supervisor. By doing so, she bypassed one of the primary timekeeping controls, which is that somebody reviews and approves the timecard before it is entered into the system.
Nice.
Ms. Holt was sentenced to four years in prison plus she was ordered to repay the $1.4 million she had obtained from her scheme.
Now that we know this is a possibility, now that we know somebody can bypass a supervisor’s review and manipulate the labor data after it has been input, how might we detect such wrongdoing?
First, let’s look at overtime. What kind of compensatory control can we devise?
We can run a report, either monthly or quarterly or annually, that shows overtime hours recorded by salaried employees (or by hourly employees). We can organize that report from highest to lowest. We can decide that anybody recording less than a certain amount of overtime hours can be dropped from the report. For example, let’s run the report quarterly and say that anybody recording less than 40 hours of overtime in a 13-week quarter will not be reported—but that anybody who recorded more than 40 hours will be reported, and that we want to see who recorded the most right at the top of the report.
Had such a report been run, we bet Ms. Holt, who was claiming roughly 625 hours of overtime per quarter, would have been right at the top.
A simple phone call to Ms. Holt’s supervisor would have revealed that the timecards that had been approved did not have such quantities of overtime hours on them … which would have led to further investigation, we presume. In other words, Ms. Holt’s scheme would have come to light within the first quarter or two of being initiated, rather than surfacing only after nearly 17 years had passed.
Now let’s talk about the Paid Time Off (PTO) manipulations.
How about we design a report that shows the PTO elements by employee, including starting and ending balances? Let’s run that report once a year.
For the average employee, there will be a starting vacation (leave) balance and an ending vacation balance. The ending balance will equal starting balance plus annual leave earned during the year, minus any leave taken. If the employee didn’t take any leave then the ending balance will equal the starting balance plus the annual accrual.
Let’s limit our report only to those employees who didn’t take any leave during the year—i.e., to those employees whose ending balance equals starting balance plus annual accrued leave. There won’t be that many employees who are on that report, because most employees take vacations. (In fact, if you work at a bank you are required to take vacations.)
For those few employees on that list, we’ll make a phone call to the supervisor to confirm that the employee didn’t take any vacation during the year. (Most supervisors know if their employees were out for a week or two; often it’s on a calendar somewhere.) If the supervisor indicates that the employee did indeed take a vacation (leave) during the year, then we’d follow up with further investigation.
Again, if we implemented that compensatory control, any shenanigans would come to light within a year or so.
Finally, holiday leave. Each company has its own holiday policy but every government contractor offers at least the Federal minimum holidays. There are 10 Federal holidays. Let’s assume eight labor hours a day. Eight times ten = 80 hours of holiday time per year. That’s what we expect to see. Sure, there are times when people work on holidays, but those exceptions are known.
We can design a report that identifies any employee who recorded less than 80 hours of holiday time per year. (You can tailor the report to match your company’s holiday policy.) There will not be very many employees listed on that report.
For each employee listed, we will make another phone call. If we learn that the supervisor was unaware an employee had worked through a holiday, we’ll investigate further.
The point is, this is not rocket science. The labor hour information you need is almost certainly in your labor accounting system. All you have to do is to design and run the reports, and then have the wherewithal to make the (few) phone calls required.
By the way, if you learn that the employee’s hours were legitimate but the supervisor was clueless about hours worked (even though that same supervisor signed-off on the timecards), then you will have learned something valuable about your primary timekeeping accuracy control. And you will also have learned something valuable about the competency of that supervisor.
We would thank Ms. Holt for prompting us to think along these lines, but that “thanks” doesn’t seem appropriate at this time. Maybe we’ll offer them in four years, or perhaps earlier if she receives time off for good behavior.