Unraveling the Peltzman Effect: The Significance of Agent’s Type

Abstract

Abstract The Peltzman effect posits that implementing safety measures incentivizes agents to reduce their effort to a degree where these measures become counterproductive. This paper emphasizes the significance of including the agent’s type (skills, attributes) when analyzing the effectiveness of safety measures. Using data from iRacing, an online racing simulator, we find that the detection of the Peltzman effect is solely attributed to the omitted variable bias; specifically, the omission of a variable capturing the agent’s type. Additionally, our data demonstrates that enhancing types (increasing skills) leads to safety improvements.