A national randomized field experiment testing the impact of simulated usage-based insurance programs on driver safety

This is the authors' abstract. We don't add key points for this paper.

Many motor vehicle crashes are due to risky driving behaviors measurable by smartphone telematics applications. These apps are deployed by usage-based insurance (UBI) programs to estimate crash risk and price policies, incentivizing customers to drive safer. A preregistered randomized controlled trial (NCT06101251) tested whether a simulated UBI program improved driver safety, and whether being assigned or choosing focused goals yielded further improvement. Adult drivers (n = 1,449) were recruited nationally with social media advertisements. After a 6-week baseline period, participants were randomly assigned to one of four groups for a 12-week intervention: 1) Observation (control); 2) Standard Feedback (weekly texts about four risky behaviors, $100 incentive); 3) Assigned Goal (weekly texts about algorithmically assigned focus behavior and goal, $100 incentive); 4) Chosen Goal (weekly texts about self-chosen focus behavior and goal, $100 incentive). Post-intervention, participants were monitored for 6 more weeks. The primary outcome was a 0-100 score for overall safety (100 = safest). Standard Feedback (covariate-adjusted difference, 2.0 [95 % CI, 1.2-2.7]; P < 0.001), Assigned Goal (2.7 [95 % CI, 2.0-3.5]; P < 0.001), and Chosen Goal (2.2 [95 % CI, 1.4-3.0]; P < 0.001) had higher safety scores than control during the intervention period and at follow-up. Differences between treatment groups were not significant after adjusting for number of comparisons, Ps > 0.10. Treatment groups had significant relative reductions in speeding (11 %-13 %, depending on group), hard braking (16 %-21 %), and rapid acceleration (16 %-25 %), Ps < 0.01. The kind of feedback and incentives offered by UBI programs can cause sustained improvements in driver safety.