Fraud in the Federal Government-Part II: Characteristics of the Schemes-The Detection and Resolution of the Cases

Article excerpt

This is the second half of a two-part article on fraud in the federal government. In Part I, published in the Spring 1997 issue of The Government Accountants Journal, the authors profiled the perpetrators and victims of 208 cases of fraud uncovered in the federal sector. In Part II, characteristics of these schemes are investigated, the methods used to detect and confirm the frauds are described and the outcomes of the fraud investigations are examined.

All data were obtained from a mail survey of the active members of the Association of Certified Fraud Examiners. Each member was asked to describe in detail an instance of fraud with which (s)he was personally familiar.' Of the usable responses, 208 described defalcations against agencies of the federal government, committed by employees and vendors, working alone and in concert.2 Specific responses were sorted by the type of perpetrator, and then analyzed using frequency statistics and averages.3

The specific illegal acts reported in this study affected numerous agencies and resulted in losses in excess of $157 million. Even beyond the sheer magnitude of the dollar losses incurred, fraud has proven to be extremely costly in other ways, such as diverting budgeted resources from intended uses, undermining an agency's attempts at reform by deflecting the focus of its leaders and eroding the public's confidence in the integrity of our federal system. Hopefully, once these fraud elements have been identified and isolated, they will be useful in the detection, identification and, even more importantly, the prevention of fraudulent activities within government.

The Schemes

While the focus of fraudulent activities may vary depending on the organizational role played by the perpetrator, the diversion of receipts and the falsification of disbursements are mechanisms frequently used to steal cash. Regardless of the scheme employed, however, collusion often prolongs the period of concealment The methods used to implement defalcations and the impact of these frauds on the victim agencies are reflected in Table 1.

The magnitude of the loss incurred by the victim agency increases with the power inherent in the role played by the perpetrator. Management ($97,200 median loss) inflicted greater damage than nonmanagement ($45,000 median loss), particularly considering the relative sizes of the organizations in which they were employed. Vendors ($400,000 median loss) and those involved in collusion ($500,000 median loss), were able to cause the greatest harm, for a number of reasons. A vendor may not have been subject to as much scrutiny as an insider working alone would have been. Thus, vendors may have been able to disguise large fraudulent dollar amounts as legitimate transactions. The cooperative efforts of colluders may have created an impression that controls were being followed, or the perpetrators may have corroborated one another's fraudulent explanations for unusual activities, again creating an environment in which large losses could occur.

The duration of employee and vendor frauds varied positively with the dollar losses incurred. Although all schemes remained undetected for a relatively long period of time, vendor schemes remained undetected for three years, while employee schemes lasted approximately two years. The duration of collusive efforts was closer to that of employee frauds (27.4 months), yet collusion was the most costly to the victim entity in dollars lost. This difference highlights the havoc that a group of perpetrators can create when controls are circumvented. Respondents judged the collusion schemes to be the most complex (4.11 on a scale of 1 to 7) of all the illicit endeavors.

Both the choice and number of the accounts targeted in fraud varied, depending on the perpetrator's ongoing activities. Nonmanagement personnel had access to fewer accounts than other perpetrators did. It was noted that 1.7 accounts were manipulated in the average nonmanagement fraud, versus 2. …