The Austin Police Department‘s second go-round with reporting racial profiling data – as required by state law, passed in 2001 – offers few surprises or standout statistical changes, and thus keeps alive arguments over disparate treatment of minorities and questions about effective reporting practices.
Of course, APD Chief Stan Knee would like to keep those arguments at bay – especially in light of the January dustup over the Austin American-Statesman‘s statistically squishy analysis of the department’s use of force. “We are deeply concerned about the interpretation and use of statistics to label police departments or create an environment where the community is polarized by such interpretations,” he wrote in a March 1 memo to mayor and council. As a result, he wrote, the department is “moving forward” with a plan to bring to town experts in “criminal justice data collection” to “validate” the department’s approach to data collection. How fruitful that plan will be remains up in the air; still, the department’s new data includes several notable findings that don’t need “expert” intervention for interpretation.
Overall, the number of APD traffic and pedestrian stops increased last year over 2002 – traffic stops rose 6%, and pedestrian stops rose nearly 21%. As with the 2002 numbers, the racial breakdown of subjects of traffic stops roughly mirrors the city’s overall population. Whites accounted for almost 52% of all traffic stops, Hispanics for 31% and blacks for just over 13%. However, the city’s “Fair Road Standards” benchmark estimates the percentage of black and Hispanic drivers at 8.3% and 20.1%, respectively, thus suggesting that minority drivers are more likely to be stopped.
The racial breakdown of the pedestrian stop data is also similar to 2002, with blacks accounting for a disproportionate 27% of all ped stops. Still, city officials (and some department critics) caution against making too much of that number, since the majority of those stops are conducted by walking beat officers working out of the Central East, Downtown, North Central, and South Central Area Commands – the areas of the city with the most foot traffic, and areas where the ethnic demographic differs significantly from the citywide average.
It’s because of factors like these, which can influence any statistical analysis of the basic traffic and pedestrian stop data, that both police officials and police watchdogs mine data subsets to look for clearer indications of race-based policing. And within those subsets of the APD data, critics say, there are clear signs of a problem. The most obvious and nearly irrefutable, says ACLU of Texas Executive Director Will Harrell, is in the percent of stops that end in searches and, in particular, stops that end with a consent search. A consent search, used when police may have “reasonable suspicion” of some illegal activity but not enough information to rise to the level of a “probable cause” search enforceable by a warrant, is by nature a tool with substantial potential for abuse.
In 2002, 7% of all traffic stops ended in a search, and just over 11% of those were consent searches. In 2003, the overall number of traffic-stop searches rose to just under 9%, with consent searches accounting for nearly 14% of those searches. But among blacks, consent searches accounted for 18% of searches – considerably more than for whites (11.7%) or Hispanics (13.4%). “The greatest disparity is in the discretionary searches of African-Americans,” Harrell said. “The obvious answer [to eliminating the disparity] is to prohibit, or create restrictions on, consent searches.”
City officials agree that the disparity is disturbing, especially considering that nearly 86% of all consent searches (regardless of race) end without officers making any sort of seizure. Still, Assistant City Manager Laura Huffman said that it is unlikely that APD will simply do away with consent searches. “Officers feel like it is an important tool for them. They feel like if they don’t have access to that tool that they don’t have a way to [act on] reasonable suspicion,” she said. “But, clearly, the data shows disparity in searches.”
But, Huffman says, it is a disparity that the department is already committed to addressing. Indeed, reducing the number of consent searches is key to the departmental “action plan” Knee announced during a late January press conference (one of the three prongs in the “Action, Accountability, Access” plan outlined by city leaders and intended to address police-community PR problems). Knee’s plan calls for a 40% reduction over the next two years in the number of consent searches that “do not yield evidence,” a feat that will be accomplished, in part, by a new, more strict policy that requires officers to document every consent search and the “legal threshold” for each search. (According to Knee’s timeline, the new reporting system should be up and running this month.)
Knee’s plan’s chances for success will hinge on the quality and comprehensiveness of the reporting – not exactly the department’s strongest asset. While the percentage of data missing from this year’s report (officially recorded as “left blank”) has decreased, the amount of missing data could be enough to skew the numbers in many cases. For example, in nearly 45% of all pedestrian stops, the official reason for the stop went unrecorded. In 44% of all pedestrian encounters there is no record of either race or gender of the person stopped. APD spokesman Kevin Buchman says that there are a “variety of reasons” that the data is incomplete, including officer error, but that the department is continually striving toward generating the most “completely completed” data possible.
Huffman says that part of that mission means always thinking about ways to “redefine data collection.” In addition to the obviously missing data, she said, there are other areas where more detailed reporting could enhance the overall understanding of how policing works – or doesn’t – in Austin. Among the reporting categories that could yield useful analysis is the “field observation report,” which officers fill out any time a stop – traffic or pedestrian – ends without any reportable action, such as a ticket, warning, or arrest. Officers record not only the race of the person stopped and the location of the stop, but also the reason for the stop and whether the person was searched. Last year, more than 29% of all pedestrian stops resulted in a field observation report; but the details related to those stops haven’t been broken out for closer observation. That data could prove important to determining whether officers use the report for potentially race-based stops, says Harrell. “To me that’s absolutely some of the most important data we can get hold of,” he said.
And, on this point at least, Huffman is in agreement. Racial profiling “is a complicated issue,” she said, “and anything that can help us sort through the cause-and-effect relationship is valuable to us.”
This article appears in March 26 • 2004.