Today brings us this headline in the New York Times:
Election Day Voting in 2020 Took Longer in America’s Poorest Neighborhoods
This is true—but just barely. In fact, I was struck by how the data actually indicates that we could be making real progress. First off, here’s a Times chart that I replotted to remove the bias of cutting off the y axis:
From the lowest to highest incomes, there’s surprisingly little variation in the percentage of people who have to wait more than an hour to vote. It basically ranges from 14 percent to 16 percent, which is barely noticeable.
Here’s another chart. It’s a little hard to figure out, so bear with me:
The gray curve represents white neighborhoods. There’s a 10-minute wait in about 4.5 percent of them, a 30-minute wait in 0.8 percent, etc. The black curve represents non-white neighborhoods. There’s a 10-minute wait in about 3.5 percent of them, a 30-minute wait in 1 percent of them, etc. As you can see, the curves are quite similar, showing only a small difference between white and non-white neighborhoods. And if you go out to the extreme, at a wait time of two hours or more, they’re identical.
I’ve drawn dashed red lines where the curves cross the 50-minute mark, which is a long-but-not-extreme wait time. In “overwhelmingly white” neighborhoods, where the red line crosses the gray curve, 0.4 percent of voters had to wait 50 minutes. In “overwhelmingly nonwhite” neighborhoods, where the red line crosses the black curve, the number was 0.5 percent. Again, these are tiny numbers and tiny differences even though the chart only includes “overwhelmingly” segregated neighborhoods.
If the Times analysis is correct—and I have my doubts—the real story here is that we have nearly eliminated the long wait times for both poor and POC neighborhoods. Anecdotal evidence makes me doubt this, but I’d sure like to see someone follow up on this using different methodology. Have we really made this much progress in polling place discrimination?