Traffic Safety Impacts of Reduced Police-motorist Interactions

Introduction

In the wake of changes to the way many police departments enforce traffic laws, there are concerns about potential impacts to road safety. The primary objective of this study was to examine measurable effects of reduced police–motorist interactions on key traffic safety outcomes. Jurisdictions have increasingly adopted policies that limit traffic stop interactions between law enforcement and motorists, typically involving limiting or prohibiting pretextual stops or stops for minor, non-safety related violations that are used to investigate more serious crimes. A policy to deprioritize pretextual stops is typically associated with an increased focus on public safety, by prioritizing violations associated with dangerous driving behaviors such as speeding, distracted driving, or impaired driving. Existing studies examining the impact of policy change on safety outcomes are limited in number, scope, and quality, but the data that are available suggest that policies reducing police–motorist interactions do not have a detrimental impact on public safety. However, most policy changes are fairly recent, and therefore the full impact of the changes may not be apparent yet. In addition, many jurisdictions have seen reduced police–motorist interactions due to a variety of factors—in the absence of a policy change.

Methodology

The current study sought to examine the relationship between reduced police–motorist interactions and safety outcomes. Six jurisdictions were selected for study. Three sites (Nashville, TN; Philadelphia, PA; and Minneapolis, MN) were selected because of a large and sustained decline in police interactions with motorists. In contrast, the other three locations (Charlotte, NC; Chicago, IL; and Saint Paul, MN) were selected because they exhibited less change in traffic stops.

Data on crashes of various severity were obtained for each site. An interrupted time series analysis was used to evaluate variations in crash trends and severity levels after a change in the number of police interactions with motorists. Monthly crash counts were analyzed using models which account for patterns like seasonal cycles and long-term trends across different severities of crashes. The impact of the reduction in police stops on speed-related crashes at all crash-severity levels was also analyzed. Lastly, rate ratios were calculated to quantify the effect size of the changes in traffic stops based on the interrupted time series models.

Results

Results on the relationship between police–motorist stops and safety outcomes were inconsistent. For two of the three locations with the biggest and most sustained declines in police–motorist stops, there was some indication of increased crashes, particularly among more severe or speed-related crashes:

• Minneapolis exhibited significant rises in fatal and serious injury crashes and known-injury crashes; results also suggest there may have been a significant increase in speed-related crashes.
• Philadelphia demonstrated significant increases in fatal and serious injury crashes and speed-related crashes, but also saw slight reductions in known-injury and total police-reported crashes.
• The third location, Nashville, however, did not see significant increases in fatal and serious injury, known-injury, or speed-related crashes; there was a small increase in total police-reported crashes. Notably, of the three sites with steeper declines in enforcement, Nashville was the only jurisdiction whose entire study period pre-dated the COVID-19 period.

The jurisdictions with more modest changes in enforcement also did not show a consistent pattern on crash outcomes:

• Charlotte showed a small but statistically significant increase in speed-related crashes and a small but statistically significant decrease in police-reported crashes.
• Patterns in Chicago indicated a significant increase in fatal and serious injury crashes and a decrease in all police-reported crashes.
• Saint Paul exhibited a significant increase in fatal and serious injury crashes, but a small but statistically significant decrease in known injury crashes and speed-related crashes.

Due to methodological challenges in evaluating the impact of changes in police–motorist interactions on road safety, it was not possible to completely disentangle the impact of changes in enforcement levels from the impacts of other factors that transpired around the same time (such as the COVID-19 pandemic). Given the difficulty of isolating the effects of changes in police–motorist interactions from other factors with variable and potentially large influence on safety outcomes, the long-term impacts of these types of changes are difficult to assess. This study’s mixed results across the selected sites do not demonstrate conclusively that reductions in police–motorist interactions impact crash rates. However, the results suggest that increases in more serious and speed-related crashes coincided with the timing of reductions in police–motorist interactions in some case-study cities.