Avoiding payroll fraud

Everyone is familiar with the phrase “dead horses on the payroll”. It describes one common form of payroll fraud, creating and paying employees who don’t actually exist. Payroll fraud is a serious problem in Canada, although there don’t seem to be any comprehensive statistics on its scale.

So how can employers work to prevent payroll fraud? Among very small employers, the most effective form of prevention is for the business owner to maintain direct control over payroll. This works, so long as the entrepreneur has personal contact with everyone who should be paid. However, at some point that becomes counter productive. Direct control by the business owner takes up too much time, interfering with running the business itself.

Failing such direct control, employers must delegate payroll responsibility to others. Once the business owner no longer has hands-on management, what can employers do to prevent payroll fraud?

There are really two very different approaches commonly taken to prevent or reduce payroll fraud. The first focuses on policies and procedures that impose accounting controls. The second looks at testing and sampling techniques. Both are intended to reduce fraud. The first by making fraud more difficult to accomplish. The second, by making it easier to find. While it might not seem as effective to focus on identifying fraud after the fact, the rationale is that exposing fraud after the fact acts as a deterrent.

These two very different approaches can be seen if we look at timesheet processing.

One form of accounting controls is to divide up responsibility for timesheets among several different people. For example, one person is given the responsibility to schedule employee time, while another person processes the resulting timesheets. Exceptions from scheduled time then require explicit supervisor approval, either in advance or after the fact. This approach can be taken, whether scheduling and timesheets are electronic or done on paper.

The assumption behind such controls is that for fraud to occur, there has to be collusion between the different people involved. For example, if employees submit claims for overtime they did not work, the rationale is that this would be harder if supervisors have to sign-off on all exception time.

However, such policy and procedure based controls have their limits. The more that supervisors and line managers are required to approve run of the mill, everyday transactions, the greater the risk that such approvals will be given without any real consideration. In other words, supervisors may simply sign off as a matter of course, without giving timesheets any real thought. The other risk is that such approvals will be forged. This is a risk, even if time and attendance systems require electronic approvals. Just as some employees may be able to forge supervisor signatures, others will be able to steal supervisor passwords. And there is always the possibility that their will in fact be collusion between the different people involved, such as a supervisor and employee colluding to split fraudulent gains.

The other approach to fraud prevention is based on testing and sampling techniques. Such techniques can’t directly prevent fraud, but should be able to discover it. The more publicity surrounds such discoveries, the more will testing and sampling act as a deterrent.

There are several ways that testing and sampling techniques can be applied to timesheets. For example, organizations can review exception pay records for unusual patterns, such as overtime hours claimed in an otherwise slow period or an uneven distribution of overtime claims between employees. Where employee time is job costed, cost accounting reports should be able to identify jobs with an unusual or high pattern of labour costs. Similarly, the accounting system should be able to identify cost centres with high labour costs or where there are large variances from budgeted labour costs.

However, looking for unusual patterns in timesheets or overtime claims may not be much help where fraud is so pervasive among employees or has gone on for such a long term that fraudulent claims themselves have become part of the normal pattern of events.

In the end, organizations have to weigh the costs any preventive measures described above against the actual losses experienced from payroll fraud.


AlanMcEwen is a payroll consultant and freelance writer with 20 years’ experience in all aspects of the industry. He can be reached at armcewen@cogeco.ca, (905)401-4052 or visit www.alanrmcewen.com for more information. A slightly different version of this article was first posted to the Canadian Payroll Reporter web site, on August 21.



About Alan R. McEwen

HRIS/Payroll consultant and freelance writer
This entry was posted in Payroll fraud and tagged , , , . Bookmark the permalink.

3 Responses to Avoiding payroll fraud

  1. We were once brought in after the payroll administrator had been caught in a serious fraud. She would create fake employees, change the bank account number to hers, and then after the payroll deactivate the employee and put the bank account back to what it was previously, so her account number was no longer on record. The software they were using did not keep track of the bank account number being changed.

    This went on for some time until the owner got to wondering about one of the employee names he saw on payroll several times, but had never heard of before…

    • Darryl, you hit the nail on the head. Smaller employers need to pay careful attention to employee names on the payroll.

      • These guys were right in the “no-man’s land” between big and small. It was a large restaurant , around a 100 employees, hard for the owner to keep track and until then, no real systems like those as you suggest. They now have systems after this lesson!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.