From the time of Hammurabi through the modern day scandals of Enron’s collapse and the Fannie Mae/Freddie Mac takeovers, fraud has been a concern. Broadly conceived, fraud is the misappropriation of assets as well as financial statement fraud (Golden, Skalak, & Clayton, 2005). The first of these – the misappropriation of assets – includes the unauthorized consumption, or theft, of an organization’s resources. In contrast, financial statement fraud is the presentation of knowingly false financial reports wherein financial damage results from reliance on those reports (Skalak, Alas & Sellitto, 2005). Accordingly, this issue has broad appeal to managerial accountants who are concerned with fraud from the perspective of control system design, to auditorswho are interested in fraud from a detection viewpoint, and to financial
accountants who are concerned with the quality of the financial reporting on which organizational and investor decisions are based.
A recent survey by the Association of Certified Fraud Examiners (ACFE, 2008) estimates losses due to all frauds at $994 billion annually.2 In addition to being costly, we know that the manipulation of financial reports also is common (Merchant & Van der Stede, 2007).
In this study, we tested hypotheses relating compensation type (contingent versus non- contingent) and the form of the contingency (bonus or penalty) on both fraudulent reporting and the misappropriation of assets. As our results suggest, the use of contingent contracts to mitigate the moral hazard of agency theory concerning the mis-alignment of interests may be short sighted in that such contingencies may encourage behaviors such that the cure itself may be worse than the disease.
Organizations have long used incentives as a mechanism for both for aligning the interests of managers (e.g. CEO) and of workers with the interests of the company and its shareholders. For example, in the 1980s, CEO Roger Smith introduced performance contingent pay to the line workers at GM – when GM did well, the workers did well (Business Week, 19xx). Similarly, performance contingent pay has long been a bastion of compensation for CEOs and top managers. Yet scholars, journalists, legislators and practitioners are among those who believe that incentives, such as contingent compensation, may actually stimulate financial misdeeds. Many have speculated on the role of incentives in the frauds at firms like Fannie Mae, Enron, Global Crossing, HealthSouth, Worldcom and Tyco. Long before these failures, well-known investor Warren Buffett, in his 1998 Letter To the Shareholders of Berkshire Hathaway, asserted that a growing number of managers manipulate earnings to inflate stock price and, therefore, personal income (Buffett, 1999). In testimony to the U.S. Senate, former Federal Reserve Chairman Alan Greenspan (2002) asserted that managers had “incentives to artificially inflate reported earnings” at least in part because their pay depended on it. Consequently, it is possible that a mechanism designed to align incentives between CEO and shareholders or the company and workers, actually results in greater misalignment.
Compensation contracts are a key element in the design of managerial control systems. Accordingly, managerial accounting research has a long history of exploring how incentives influence behavior (for reviews, see Luft & Shields 2003; Bonner & Sprinkle 2002; and Young & Lewis 1995). For example, the literature on participative budgeting has found that managers create budgetary slack by requesting more resources than are needed, or by understating the ability to produce. Examples include Young (1985), Waller (1988), Chow et al. (1988), Chow (1991a) and Chow (1991b). This perspective is similar, yet somewhat different from our notion of fraud. More closely aligned with our study is the work of Baiman & Lewis (1989), Waller and Bishop (1990) and Evans et al (2001). These studies have focused on misreporting, finding that compensation does lead to greater opportunism. These studies are discussed in greater detail in the next section, when we develop our hypotheses.
Financial accounting researchers have also looked at this issue in the sense that financial fraud is one aspect of earnings management. A less extreme form of earnings management is aggressive accounting (Dechow & Skinner, 2000). Schipper (1989) provided an early definition of earnings management as “‘disclosure management’ in the sense of a purposeful intervention in the financial reporting process, with the intent of obtaining some private gain” (p. 92)3 (for reviews, see Schipper  and Healy & Wahlen ). Briefly, the earnings management literature generally focuses on one of three topics, total accruals, specific accruals and frequency distributions of earnings (McNichols, 2000). In an early study, Healy (1985) used total accruals as a measure of managerial discretion, and found evidence that managers with a bonus cap deferred income when the bonus cap was achieved in order to maximize future personal earnings. As the literature evolved, research focused on specific accruals and frequency distributions of earnings as a way to identify earnings management (e.g, Burgstahler & Dichev, 1997). Healy & Wahlen (1999) concluded that market expectations, management compensation and regulatory intervention were the principal factors motivating earnings management. While his literature has found compensation contracting to be a principal source of earnings management, this stream of research has largely ignored the specific causal mechanisms.