Mitigating Gig and Remote Worker Misconduct

Jobs in which workers are physically distant from their employers are increasingly prevalent, due to both a surge in the gig economy and the widespread increase in remote work, which was on the rise even before the pandemic. This development has created unique employee-governance challenges.

Our forthcoming article, “Mitigating Gig and Remote Worker Misconduct: Evidence from a Real Effort Experiment” posits that employee misconduct will likely be prevalent in gig and remote work settings because the physical separation between employers and workers exacerbates the “principal-agent” problem in two important ways. First, gig and remote workers are likely to feel less connected to their employer,  increasing the separation between their interests and the firm’s. Second, physical separation typically comes with more information asymmetry. Under these conditions, employee misconduct – already costing U.S. firms as much $600 billion annually in traditional work settings  – is likely to be a critical and growing challenge.

Our article proposes a way for employers to potentially mitigate gig and remote worker misbehavior: through communication of  organizational values, which we argue increases the alignment of interests between worker and employer by fostering a sense of shared values. Next, since monitoring gig and remote workers is often challenging, we consider how the credibility of the threat of monitoring is likely to affect misconduct in these settings. Finally, we also predict that policies emphasizing organizational values are likely to be less effective when monitoring (or the threat thereof) exists. Specifically, we argue that the threat of monitoring is likely to lower perceived trust between worker and employer. This inhibits workers from forming the sense of shared values with the employer that the communication of organizational values would otherwise elicit – thus reducing the effectiveness of this approach in aligning workers’ interests with those of the organization. This crowding out effect is likely to apply not only to gig or remote work settings, but to traditional work settings as well, and to our knowledge it has not been theoretically examined or empirically tested in either.

Assessing the effectiveness of employer-level policies intended to reduce misconduct is empirically challenging for two reasons. First, misconduct is exceedingly difficult to measure accurately in real work settings since workers tend to hide it. Second, employer-level policies such as those outlined above are rarely exogenous in practice, making it difficult to separate their impact on misconduct from that of other factors. To sidestep these challenges, we designed a novel experiment in a gig work context which enabled us to observe and accurately measure misconduct.

In our study, workers were hired on an online gig-market platform to complete a short-term assignment of entering information into the “Contact” sections of various websites. They could also earn bonus payments by contacting the website owners by phone, with instructions to either leave a scripted voicemail or to obtain answers to a market research survey. Unbeknownst to the workers, we owned and operated both the websites to which they were directed and the corresponding phone numbers listed. As a result, we were able to cleanly observe whether workers actually entered the requested information into the website as directed and  whether they fraudulently claimed any bonus payments from their employer.

Workers were randomly assigned to one of six conditions in a three-by-two design. First, workers randomly received a message about the employer’s ethical values (an expression of internally-oriented organizational values), a message about the employer’s social and environmental values (an expression of externally-oriented organizational values), or no additional messagies. Second, we manipulated whether or not workers were told that their work might be monitored. Notably, from the workers’ perspective, the credibility of the monitoring threat varied between the primary task (where the threat was not particularly credible) and the bonus task (where the threat was at least partially credible) – allowing us to shed light on the extent to which credibility matters.

We find that, when implemented independently, the communication of organizational values and the threat of monitoring are effective at reducing employee misconduct by statistically and economically significant degrees. (Indeed, even when not credible, the threat of monitoring reduces misconduct – though the effect is less robust.) When implemented in combination, however, the effects of values-oriented policies and monitoring policies are not as effective as one might expect.  Specifically, when the threat of monitoring is in effect, the communication of organizational values does very little to further reduce misconduct. We explore potential underlying mechanisms and find evidence consistent with the theory that monitoring erodes the trust needed to establish a sense of shared values between workers and the firm. Indeed, empirical results indicate that the threat of monitoring substantially reduces workers’ perceptions of employer trust. And the communication of ethical and social and environmental values increases workers’ perceptions of shared values with the firm – but only when monitoring is not also in effect.

This post comes to us from professors Vanessa Burbano amd Bennett Chiles at Columbia Business School. It is based on their recent article, “Mitigating Gig and Remote Worker Misconduct: Evidence from a Real Effort Experiment,” available here.