Racially-biased policing in Oakland (updated with 2019 data)

The Oakland Police Department is legally required to provide racial data on police stops. But they’re not required to make it easy. But they finally released their report.

We don’t have detailed public data on where the stops are occurring, only the police beat. I did some hacky stuff to estimate the racial composition of each police beat using Census data, and tested whether the stops were proportional to the racial makeup. As you might guess, the answer is “no.” In 2019, the problem was marginally better, but it’s still a problem.

Collection of Chicago (North Lawndale) maps

I’m working on a series of maps of Chicago’s North Lawndale neighborhood. I’m just going to drop a bunch of drafts into this post to share with the team.

Map of Chicago, showing Burgess' concentric-circle zone model on top of the FHA redlining map for Chicago. Nearly all of zone 2 (zone in transition) and zone 3 (zone of workingmen's homes) were redlined. Zone 4 (residential zone) is mostly yellow, and zone 5 (commuters zone) is mostly green and blue.

Burgess concentric circle map in GIS

I’m going to be doing some mapping for a project from Equiticity, and one of the themes will be historical spatial inequities in Chicago. This got me thinking about the highly influential concentric-circle city development map drawn by Ernest Burgess (Chicago School of Urban Sociology) in 1925.

Surprisingly I couldn’t find a usable GIS representation of his drawing. So I decided to work on my georectifying skills and put one together.

You can see how Burgess’ racist ideas led directly to racist housing policies.

Map of Los Angeles County census tracts, colored by percentage of Black population, overlain with Metro Bike Share system coverage.

Assumption of equity

I see advocates assume that the projects that they advocate for will address issues of historical inequities. A thread on distributing bikes in LA led me to investigate whether LA Metro bike share is equitably distributed. Bike share does not reach the Black areas of the city at all; in fact, there’s not a single bike share station located in a census tract that is even 25% Black.

Displacement and density

After doing some crunching this week on data about rapidly-gentrifying Valencia Street in San Francisco, and finding that residential density is actually dropping in the neighborhood despite new housing construction, I wondered whether the same phenomenon could be found elsewhere in the country. I didn’t have to wait long for another case study, as Lynda Lopez and a number of other peeps I follow from Chicago posted about Mayor Lori Lightfoot’s ill-considered statement about “vibrancy” in Pilsen, a gentrifying neighborhood west of the Loop. It turns out that the demographic dynamics in Pilsen are very similar to Valencia Street, and they challenge the claim that infill development is a climate solution.

Two maps of Valencia Street in San Francisco, showing a large increase in median household income from 2000-2018, and a sizable decrease in Hispanic population.

Bike infrastructure and displacement

I have been really, really trying to avoid getting into the housing debate. But transportation and housing are inextricably linked, and my investigation of bike infrastructure’s relationship to gentrification is part of the Bike Lab’s work. A Twitter thread sparked by Phil Matier’s terrible SF Chronicle article led me to do some data analysis of Valencia Street’s transformation. Some of the results are pretty staggering.

Map of San Francisco showing Golden Gate Park and a selected census tract (67000) in the Outer Sunset

Error bar

Phil Matier is a terrible journalist. One of the ways in which he’s terrible is that he doesn’t understand statistics at all, or at least, the way he uses statistics in his writing is uninformed (at best) or intentionally misleading (at worst). I’ll be charitable and assume that the fallacies in his work come from a place of ignorance rather than deception.

Chart showing fatal crashes in Vision Zero cities from 2011-2018, gradually rising about 20%.

Revisiting a bad idea

My posts earlier this year suggesting that Vision Zero’s communications strategy of amplifying safety risks might be counter-productive received a little bit of attention. I wanted to revisit the analysis to include the 2018 data from both the ACS and FARS, and to address some methodological criticisms.

Biking while Black part 2: Racially-biased policing in Richmond, CA

Jesus Barajas suggested to me that Allwyn Brown in Richmond (CA) had done a lot of work to change the culture of RPD. And since I’m going up to Rich City Rides tomorrow, I thought I’d see if I could run some numbers there. Getting good data is hard, but I was able to find some sources. The story is, unfortunately, no more encouraging.

Map of Oakland, displaying the large number of stops in West and East Oakland, and the generally large proportion of African-Americans stopped in all police beats

Biking while Black: Racial bias in Oakland policing

After his arrest, Najari asked me to help out with research and data analysis on racially-biased policing in Oakland. Since then I’ve also been providing information to the team working on Let’s Bike Oakland, the bike plan update, and found that 68% of bike/ped stops in Oakland were of Blacks, who comprise 24% of the population. OPD argues that the bias is spatial rather than racial; they police more in Black neighborhoods because that’s where the crime is. Does that claim hold up under scrutiny? It turns out the answer is no: Blacks are significantly more likely to be stopped no matter where they are in the city.

More on a bad idea

My post on Vision Zero’s threat amplification communications strategy got some interesting responses. One class of response as a sigh of relief, from advocates and city officials who don’t want to adopt such a confrontational style. Another was an interest in comparison data; what happened in non-Vision Zero cities over the same time frames? An entirely reasonable question, and easy for me to research.

Three maps, showing non-white population in Oakland along International, MacArthur, and Skyline Boulevards.

MLK Way: Conclusion

The biggest lesson I take from this project is that urban Black communities (and disadvantaged communities in general) have complex challenges, and those of us who care about equity and social justice need to grapple with that complexity. “Gentrification” is a reductive term which avoids meaningful engagement. While all of this is definitely Black History, it’s also White history. Those of us who believe in social justice as a concept, and who have benefitted from racist policies advantaging us and our families, need to learn to participate in social justice as a practice.

Scatter plot, displaying a correlation between increase in White population and increase in income

MLK Way part 9: Summary data

If you prefer charts to maps, here’s the post you’ve been waiting for; aggregate data for all 58 study cities, with bar charts, scatter plots, sums, medians and correlations. Woot! Interestingly, a number of my field work cities show up prominently in the data.

MLK Way part 8: Obligatory bike content

The Bike Lab began as an attempt to investigate the chicken-and-egg question of whether bike lanes led to gentrification, or gentrification led to bike lanes. In the end I found that the more interesting question was why we came to associate bike lanes with gentrification, given that the strongest predictor of urban cycling in the U.S. is being a low-income ethnic minority. But I can’t do a series on neighborhood change without talking about its relationship to cycling rates.

Map showing increase in Hispanic population in San Antonio, by census tract, with northeastern tracts much greater than western tracts

MLK Way part 7: Shifting poverty

To this point I’ve been mentioning only White and Black populations, but the most substantial demographic shifts nationally are among Hispanic populations. There has been a net influx of international Hispanics, and the natural population growth rate is also higher than Whites and Blacks.

As a population, Hispanics are wealthier than Blacks but still far less wealthy than Whites. In many cities, Blacks and Hispanics are now competing for whatever inexpensive housing exists. In these seven study cities, this manifests as an increase in total population, combined with a drop in Black population, a rise in HIspanic population, and a drop in real income.

Map of southeastern Washington DC showing changes in demographic composition from 2000-2017

MLK Way part 6: Development without displacement

The Holy Grail of community economic development is “development without displacement”: reinvestment in decayed neighborhoods which can provide long-time residents with new opportunities, without forcing them from their homes. It’s hard to achieve in a market-driven society, especially one where home ownership rates and household wealth are as wildly disparate as they are here in America. But it’s not impossible; six cities in the study qualify.

MLK Way part 5: Resegregation

On the eve of the Super Bowl, CityLab posted an article about a neighborhood in Atlanta which was destroyed to build the Georgia Dome, which itself was later destroyed to build Mercedes-Benz Stadium, where the game was played. It so happens that Atlanta’s Martin Luther King, Jr. Drive runs right by the stadium. It being Black History Month, I thought I should dig into the history of the place.

Maps showing Oakland census tracts intersected by MLK Drive in 2000 and 2017

MLK Way part 4: Displacement

The Black Panther Party got its start on Oakland’s Martin Luther King, Jr. Way (then named Grove Street). It was partly fear of the Panthers which led the city to make Grove/MLK the boundary where the elevated BART tracks and the 980/24 freeway would be routed. Anyone living there today can tell you that the neighborhood is one of the most rapidly gentrifying areas of Oakland, which is why it shows up in the group of cities where MLK Way is undergoing displacement: a decrease in Black population and an increase in White population coupled with an increase in median income and other markers of socioeconomic change. This is the most common case in the data set, with 15 cities meeting the criteria.

Map displaying Cleveland census tracts intersected by Martin Luther King Drive, by racial composition

MLK Way part 3: Displacement without development

The most obvious failure of policy is in those neighborhoods which are experiencing “displacement without development”: decreases in both population and income. 12 of the cities in the study fit this profile.

Side-by-side maps of 2000 and 2017 Census boundaries, showing slight differences

MLK Way part 2: Caveats about data

I’m going to be throwing a lot of data out there in this series, so I wanted to start with a few caveats about my process, and about data analysis in general.

Accurately describing the social dynamics of the city through data analysis is a persistent challenge, and the geospatial dynamics aren’t much easier to get at. A simple-sounding question like, “how has the proportion of Blacks in this neighborhood changed over time?” runs into all kinds of problems with the definition of terms and the availability of data.

Scroll to Top