Yesterday, I discussed the impact of police deployment practices on the racial disparity apparent in traffic stops by the Lincoln Police Department. Today, I am focusing on the Big Dog: income. More than any other factor, racial differences in income influence disparity in traffic stops.
Back in 2003, I wanted to explore this possibility. I obtained an export from our records management system of all traffic tickets written by LPD officers the preceding year, 2002, then dropped those data into an Excel spreadsheet. Using pivot tables, I produced counts and queries by offense type. The racial disparity for some offense types, such as driving while suspended, fictitious plates, no insurance, and improper registration, was exceptionally large: a factor of four or five for African-American motorists compared to population.
For the most common traffic offense, speeding, there was no racial disparity in the 2002 traffic citation data. The most disparate offenses were all related to one's ability to buy insurance, pay registration fees and taxes, pay fines, reinstatement fees, and the like. These income-related offenses were driving the overall disparity. I did the same thing the following year with 2003 data, then again last week with the over 50,000 traffic charges from 2013. Basically, the same results emerge by offense type, looking at tens of thousands of tickets.
I also did some work using the field in the data for the model year of the vehicle. I reasoned that the model year of the vehicle might be a rough indicator of income, and my thought was to use it as a stand-in variable. When you look at late-model vehicles--those in the most recent five model years--racial disparity pretty much evaporates.
I think these findings are intriguing, and suggest a significant influence of income on racial disparity in traffic stops. Research Methods was one of my toughest courses in grad school, and I'm no PhD. I have discussed my ideas and offered these data for a decade to several people who are, however, including faculty members in the University of Nebraska at Omaha School of Criminology and Criminal Justice, from which I am a graduate.
Every academician I've talked to about this has been interested in what I have found, but so far there have been no takers. Last week, after reading the news that Attorney General Eric Holder announced a major Federal research effort on racial profiling, I contacted the portfolio manager for that research at DOJ with yet another offer. We shall see. All law enforcement agencies collect the offense type and the model year of the vehicle on their traffic tickets. The work I've done in Lincoln could be repeated almost anywhere, and a broader sample might provide more insight into how income affects racial disparity in traffic stops.
Here's why this is important: if most of the racial disparity is the result of income differences, we are barking up the wrong tree trying to fix it with the usual formula: screening police applicants for signs of racism, implicit bias training for officers, enhanced accountability systems, and so forth. This doesn't mean these steps have no value (they are valuable in their own right), but they are not likely to impact disparity because they are not related to its cause.
It is worth noting that racial disparity in police traffic stops is only the tip of the iceberg. We live in a country where there is huge racial disparity in educational outcomes, income, incarceration, health, and even life expectancy. Deal with the first one, educational outcomes, and the others will largely disappear over time. That's what I've been advocating.
Links to the series: