Employee collusion is a significant and growing problem in organizations. It can take on many forms, from outright fraud to more subtle collaborations between employees to serve their own interests at a cost to the organization as a whole. For example, managers may join forces to game the budgeting process to realize easier performance targets and ultimately receive a higher bonus. Organizational transparency, which can increase accountability and encourage collaboration, is often proposed as a mechanism to reduce corruption and fraud.

Yet our research shows that transparency can be a double-edged sword: The more employees know about what’s going on elsewhere in the organization, the better they’ll be able to differentiate between colleagues who might be interested in joining an ill-intended partnership and colleagues who are likely to remain loyal to the company. This, in turn, makes their collusion attempts more successful. The research also indicates, however, that managers in more transparent organizations tend to treat their employees better, thus reducing the willingness of these employees to defraud the company.


The Association of Certified Fraud Examiners (ACFE) estimates that the median U.S. company loses the equivalent of 5% of its annual revenue to fraud and that half of the fraud cases involve collusion schemes. Moreover, the ACFE indicates that the median loss of collusive frauds is 259% higher than the median loss of frauds committed by a single person.

Fraud involving collusion is also more difficult to prevent and less likely to be exposed. Many anti-fraud controls work on the principles of separation of duties and independent checks, but, by working together, fraudsters can circumvent these internal controls. What, then, can companies do to prevent employee collusion?

To answer this question, it’s important to consider why employees collude in the first place: Research has shown that employees who feel they’re treated badly at work are more likely to engage in dysfunctional and fraudulent behavior (Clara Xiaoling Chen and Tatiana Sandino, “Can Wages Buy Honesty? The Relationship between Relative Wages and Employee Theft,” Journal of Accounting Research, September 2012).

We set out to investigate whether managers’ kindness is also a determining factor when it comes to employee collusion. How does an employee know whether a colleague might be interested in joining a collusive effort? We argue that they’ll consider how this colleague is treated by their manager. More specifically, we predicted that employees looking for partners in crime would single out colleagues who are treated unkindly by their supervisors.


Employees likely have some idea how a colleague from the same department is treated by their boss, but it’s more challenging if it’s someone who works elsewhere in the company. There’s much variation in how transparent companies are about what exactly goes on in different units, especially about how the managers of different units interact with their employees. While in some companies a lot of information is shared via webpages, newsletters, social events, and so on, other companies are much more closed and uncommunicative. Notably, the level of internal transparency isn’t always the result of a deliberate policy.

In more transparent organizations, ill-intended employees will find it easier to identify peers who are working for an unkind manager. These peers are likely less happy in their job and less loyal to the company and could be more willing to join a collusive effort.


To increase our understanding of the complex relationship between internal transparency, manager kindness, and employee collusion, we conducted two experiments (see “Finding partners in crime? How transparency about managers’ behavior affects employee collusion,” Accounting, Organizations and Society, January 2022).

The first experiment simulated a business setting in the economics laboratory of a large university. Groups of four participants formed an organization consisting of two departments, each with a manager and an employee. The experiment consisted of two stages. In the first stage, managers decided how kindly they treat their employee. More kindness increased the payoff of the employee but decreased the payoff of the manager.

In the second stage, employees decided whether or not to initiate collusion with the employee of the other department. Collusive agreements are made if—and only if—both employees initiate collusion. Employees incur a cost once they initiate collusion, but payoff is increased when a collusive agreement is made. We manipulated transparency by varying whether or not employees know about the kindness of their peer’s manager prior to making their collusion decision.

The results of this experiment were consistent with our expectations. Employees who know how their peers are treated by their managers are less likely to initiate collusion with peers who are treated kindly and more likely to initiate collusion with peers who are treated unkindly.

We also found that, over time, managers begin to understand that treating employees badly backfires more strongly in more transparent organizations. As a result, these managers end up treating their employees better than managers in nontransparent organizations do, leading to fewer collusion attempts.

The conclusion of this experiment is that on the one hand, we see fewer collusion initiatives in more transparent companies because managers in these companies treat their employees better. On the other hand, collusion initiatives that do emerge are more likely to succeed because it’s easier for colluders to find willing accomplices.

The second experiment consisted of a vignette study conducted online with participants with many years of work experience. The participants assumed the role of a consultant who approaches a colleague with a proposal to jointly bias a cost estimate and thus secure a lower target and a higher bonus. We varied what the participants knew about the colleague’s direct superior: nothing, that this superior is a kind person, or that this superior is an unkind person.

We found that participants who know that the colleague’s boss is kind are less likely to make a collusion proposal than participants who know nothing about their colleague’s boss. Also, participants who are informed that the colleague’s boss is unkind are more likely to collude. This is consistent with our expectations and with the results of the first experiment.

The main lesson from our research is that increasing transparency isn’t a perfect remedy when it comes to preventing employee collusion. Transparency can increase or decrease collusion, depending on what it reveals about how managers treat their employees. Thus, before opening up, companies should carefully consider what type of people typically occupy leadership positions in their organization. The good news is that, over time, transparency motivates managers to become kinder to their employees, ultimately benefiting everyone in the organization.

About the Authors