What Can Lyft Data Tell Us About Racial Bias?
My interest in racial inequities, or more generally, the demographic differences in economic outcomes, stems from my experience as an Arab living in a post 9/11 United States. Seeing how differently I was treated pre and post 9/11 helped me realize that bias exits, and racism isn’t “over”.
Bias in human interaction exists. Whether the bias is explicit or implicit, it remains present. Bias has severe consequences to economic outcomes. Debate on how much of economic difference can be attributed to bias is due to the difficulty of measuring bias accurately. However, a new paper uses high-frequency rideshare data to examine racial bias in traffic stops. This research is novel because it solves problems that other researchers in this area have never been able to.
Value Added
Previous research in this area suffers from selection bias. The data collected consist of police-reported outcomes, and those are non-random. The data only includes the people that the police chose to interact with, which is very different than the overall population. Previous studies would under estimate the amount of racial bias.
“First, there may be a selection bias in the data. Prior studies use police-reported data on drivers who experience a certain outcome, such as being issued a speeding ticket, and examine inequities by racial group in post-stop outcomes such as leniency in fines or the use of force. However, if there is non-random selection into the set of people who are stopped, such methods are likely to underestimate bias.”
The second area that this research improves upon is the accuracy of the measurement. Using police reports and speeding citations does not provide an accurate representation of the actual speed of the driver, just the reported speed. By using driver location data, the researchers are able to observe actual driving speeds and therefore avoid any measurement error.
Economics with Dr. A is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.
The Data
The study relies on several data sources
Florida Traffic Citations
Lyft location and ride sharing data between August 2017 and August 2020. This data allows the researcher to observe speeds the car was traveling at.
Florida Department of Transportation road segment speed limits. It allows the researchers to measure how much above the speed limit the driver was driving. It also allows us to observe the difference between the true traffic violation and the reported traffic violation on the traffic citation.
Findings
Minority drivers are 24 to 33 percent more likely to receive a speeding ticket for traveling the exact same speed as white drivers.
These differences amount to minority drivers paying 23 to 34 percent more in fines for the same level of speeding as white drivers
White drivers were slightly more likely to drive 0–9 mph over the speed limit.
There was no statistically significant differences for speeding more than 10 mph over the limit
Policy Implication
For governments- Relying on human enforcement is plagued with bias and leads to unequal treatment. Speeding cameras will reduce selective enforcement of traffic regulations.
For business- Insurance or employment decisions using driving records can lead to racial differences in economic outcomes because they use biased measures of risk. It will also improve their profitability if they measured and priced risk accurately.
Conclusion
This research highlights that racism, and bias, are not an individual problem. The economic outcomes we observe stem from racial differences in treatment. Ignoring this possibility will lead to biased results in research, and biased policies by government and businesses.
Other reading
