Online Labor Market Signaling with App-based Monitoring

App-based monitoring is widely used in online labor markets to reduce information asymmetry between firms (employers) and contractors. Such monitoring apps provide employers with detailed information on contractors' work progress and effort level. A unique characteristic of app-based monitoring systems is that, due to online privacy regulations, contractors are informed of their presence and explicitly agree to their installation as a condition for participating in online labor markets. This study extends the existing theory on monitoring by analyzing how this awareness of monitoring could affect contractors' strategic behavior. We explicitly recognize that app-based monitoring provides a new signaling opportunity to contractors as contractors are aware that employers can observe their behavior. We develop a new signaling model with the following features. 1) Online contractors can signal their types through wage quotes without the presence of app-based monitoring, and the app-based online monitoring can perfectly reveal the online contractor's effort level. 2) The online contractor is aware of monitoring and can signal his ability through wage and/or effort level. 3) The employer infers the contractor's type from both wage quotes and observed effort level and disseminates the information in the online labor market. 4) the employer maximizes its profit by deciding on whether to hire the contractor based on the proposed wage. We show that, by providing a new way of signaling to contractors, app-based monitoring can dramatically change contractors' wage quotes and effort level in equilibrium. In some cases, the high ability contractor proposes zero wage and the first best effort level, i.e., the effort from the complete information case, to signal his ability. We further find that, under certain conditions, the market outcome could reach the first best. However, the monitoring might distort the labor supply by incentivizing an effort level higher than the case of complete information. Several implications for platform design are generated from our equilibrium predictions