Do Electronic Signs Displaying Number of Traffic Fatalities Actually Cause More Crashes?

From Jonathan Hall & Joshua Madsen, Can behavioral interventions be way too salient? Proof from targeted visitors safety messages, in the journal Science:

Policy-makers are progressively turning to behavioral interventions such as nudges and informational strategies …. Guidebooks say that these interventions should really “seize people’s interest” at a time when they can take the wished-for action, but little consideration has been supplied to the costs of seizing one’s interest and to the possibility that these interventions may well group out other, extra crucial, factors.

We estimated these expenses in the context of a … behavioral marketing campaign with the said aim of lessening visitors crashes. This marketing campaign shows the 12 months-to-day number of statewide roadside fatalities (fatality messages) on previously put in freeway dynamic message signals (DMSs) and has been implemented in 28 US states….

We approximated the influence of exhibiting fatality messages utilizing details from Texas. Texas gives an suitable environment because the Texas Section of Transportation (TxDOT) decided to clearly show fatality messages starting up in August 2012 for 1 7 days every single thirty day period: the week right before TxDOT’s monthly board meeting (marketing campaign weeks). This permits us to measure the influence of the intervention, keeping mounted the street section, year, thirty day period, working day of week, and time of day. We utilized info on 880 DMSs and all crashes happening in Texas in between 1 January 2010 and 31 December 2017 to investigate the effects of this security marketing campaign. We believed how the intervention has an effect on crashes near DMSs as very well as statewide. As placebo assessments, we estimated no matter if the chosen months inherently vary employing details from ahead of TxDOT started out exhibiting fatality messages and info from upstream of DMSs….

Opposite to coverage-makers’ anticipations, we observed that exhibiting fatality messages boosts the amount of site visitors crashes.

Campaign months notice a 1.52% increase in crashes inside of 5 km of DMSs, a little diminishing to a 1.35% improve around the 10 km soon after DMSs. Since of imperfect compliance and competing needs, targeted traffic engineers do not display fatality messages on all DMS hours through campaign months, implying that the outcome of displaying a fatality concept on a DMS is even more substantial. We employed instrumental variables to get better the result of exhibiting a fatality message and document a substantial 4.5% maximize in the range of crashes over 10 km. The result of exhibiting fatality messages is equivalent to elevating the pace restrict by 3 to 5 miles for every hour or reducing the quantity of freeway troopers by 6 to 14%…. Back-of-the-envelope calculations propose that this marketing campaign causes an more 2600 crashes and 16 fatalities per 12 months in Texas alone, with a social price tag of $377 million for each yr.

Our proposed explanation for this shocking locating is that these “in-your-experience,” “sobering,” negatively framed messages seize also significantly interest (i.e., are way too salient), interfering with drivers’ means to respond to alterations in visitors conditions….

Thanks to Prof. Glenn Reynolds (InstaPundit) for the pointer.