
Can a shift from punitive retributive justice towards restorative justice lead to **worsening** incident performance?
This paper may interest my Aviation friends: it explores various indices of incident performance pre and post policy FAA policy changes.
Prior to the change, the FAA system relied on retributive justice, relying more on sanctions and punishment. Post change, it relied on “removing the fear of punishment to promote information sharing”, removing punishments from non-compliance (with some exceptions), and some other changes.
They evaluated NTSB data on incident counts and runway incursion from 2010 – 2020.
Findings:
· An “association was observed between the change in FAA’s enforcement policy and the increase in aircraft incidents, accidents, and runway incursions post-intervention (2015–2020)”
· And these “adverse safety outcomes appear to be directly related to this policy change”
· Though it looks like incidents were already increasing year-on-year from 2010 onwards, they observe an elevated effect from ~2015 onwards
· They found an “overall decrease in aviation safety since the inception of the Compliance Program in 2015”
· “Though the Compliance Program aims to increase aviation safety, the results show that this policy has not increased aviation safety; in fact, the results illustrate that the reverse may have occurred”
· They note an increase in incidents by 14.9%, 27% increase in runway excursions and other changes
· They also observe an immediate effect of the policy change was a “significant decrease in the number of enforcement actions (e.g., punishment) inversely proportional to the number of Compliance Actions”
· “this shift from retribution to restorative justice was deemed “very likely the direct result of the Compliance Program”

For why these policy changes may have negatively impacted the measured indices:
· They argue about a reduced deterrent effect, and hence removal of punishment may result in a “removal of an adverse stimulus, such as a warning notice, the behavior (undesired in the present case) is strengthened (i.e., negative reinforcement)”
· And some other stuff that I’ve skipped.
Of course, several notable limitations are present.
What I’m not clear about is if, and how, they accounted for the point that perhaps post-change, more people were now empowered to report incidents; which would inflate incident numbers.
Ref: Calabrese, C. G., Molesworth, B. R. C., Hatfield, J., & Slavich, E. (2022). Transportation Research Part A, 163, 304–319.

Shout me a coffee (one-off or monthly recurring)
Study link: https://doi.org/10.1016/j.tra.2022.07.016