Daily Kos's own DocDawg, aka Bill Busa, isn't new to finding and breaking voter suppression news from his state of North Carolina here. Earlier this year,
the story of voter registration irregularities under the new regime of Republican Gov. Pat McCrory. That research has been included in litigation against McRory's voter suppression efforts. Busa has been doing a lot more research, and broke the story for us last weekend, at the Daily Kos Connects Asheville Conference.
What Busa has found is illustrated in the above image: a systematic and very sophisticated effort to make the simple act of going to the polls much, much more difficult for North Carolina's black voters. From Busa's presentation:
Most of the voter suppression actions taken by the new Republican majority legislature and McCrory have been very transparent, very apparent to the public: redistricting to corral black votes; voter ID; reducing early voting days; eliminating same day registration; and, particularly damaging to the black voter, eliminating voting on the Sunday before the election, the traditional "Souls to the Polls" activity of black churches. All of that was well publicized. This, however, this eliminating and moving of polling places was done very quietly. In Busa's words: "There was no systematic 'we're going to go into one hundred counties, we're going to steal polling places from blacks in all one hundred counties.' What there was was a rather more efficient and slimy way of doing the same thing." It was done under the radar of everybody but Busa and his partner, who sniffed out the problem and have the statistical analysis chops to document it.
Watch his presentation and read the transcript of the whole thing below the fold. Get mad, then get organized. How do you get organized? By doing what DocDawg calls for:
Transcript of Bill Busa's presentation at Daily Kos Connects Asheville,
Sept. 26, 2015
What I want to do today is to persuade you that any political organizing activity whatsoever in the 21st Century is wasted unless it's supported by reasonably sophisticated data analytics. There's a world of data out there, and if we don't have that data, and we don't analyze that data, we lack situational awareness—we don't know what the enemy is doing to us, we don't know where their troops are...and we need to correct that.
The way I'm going to try to persuade you of that today is by painting a story: I'm going to tell you a story about an outrage you don't know about—nobody knew about, until my very good friend, whom I can't introduce while we're on the internet, and myself, sat down and looked at some interesting data the other day.
So this is a story about...as you know, Republicans tend to be afraid of African Americans. And, in particular, Republicans are afraid of African American voters. And in very particular, North Carolina Republicans are terrified of African American voters, because African American voters are 22% of the electorate in North Carolina. That's a big chunk of votes. And unlike white folks, black folks tend to vote their best interests.
And so, as a consequence, North Carolina Republicans have, for years now, been very effectively pursuing a multifaceted voter suppression strategy in this state—parts of which you all know about, and one part, which I'm going to show you today, that you don't know about. So let me very quickly review North Carolina's cutting-edge voter suppression program.
You all know that the Republican super-majority in North Carolina's state legislature had the benefit of presiding over re-districting in 2011, due to the 2010 census coming out. And, you know, no great surprise: they gerrymandered the bejeezus out of North Carolina, primarily to corral black voters into a very small number of very black districts, so that they could have the majority of everything else. Now, they didn't just do that for blacks, they did it for liberals, and college towns, and for other people of color as well. But I think that the biggest damage was done with respect to corralling black folks in a small number of districts. So you know about that story.
You also perhaps know about the Voter Information Verification Act, which was passed in 2013, ironically nicknamed ‘VIVA.' The screen is not large enough for me to list all of the things that this classic voter suppression act does, but just some of the highlights are here. It requires voter ID, which is very difficult for some people to get their hands on. It cut the number of Early Voting days in this state in half; that's a big deal because black people overwhelmingly use early voting—over 70% of black voters in North Carolina vote early, rather than voting on Election Day. The number is a lot less for white voters, I think it's just barely 50%. So that disproportionately impacts black voters. They eliminated same-day registration, which is a big deal if someone gets religion on Election Day and decides he wants to vote, he should be able to register that day, right? No. And, finally, one of my big sticking points is: in North Carolina, and throughout the South, where the black church is an important political power, black churches organize these ‘souls-to-the-polls' Sundays—the last Sunday before election day, after the service, everybody gets on the church bus, they drive out to the polling place, and they vote. It's a very effective way to get large numbers of black voters out. But the legislature decided in its great wisdom that we shouldn't have Early Voting on the last Sunday before Election Day. Just coincidentally, you know. Those are among the many things that they've done here.
If you ask people about voter suppression in North Carolina, those are the two stories everybody knows about, because those are the two stories that the mainstream media has actually put a little bit of effort into covering. But there are additional stories, and one that I was really lucky to be able to break, on Daily Kos, earlier this year...I was looking at some data published by the state—I look at data for fun—regarding how many new voter registrations came from social services offices in North Carolina, and from the DMV. There's this thing called the National Voter Registration Act...the NVRA, or the ‘Motor Voter' law...passed in 1993 in the Clinton administration. It requires any state agency that has face-to-face interactions with citizens to ask them—not just to have a pile of voter registration forms on the counter, but to ask them—"would you like to register to vote today?" It's a legal requirement. And what I found while staring at these rows of numbers was that, immediately at the beginning of the McCrory administration, when a Republican governor came into office and took it over from the previous Democratic administration, what had been a couple of thousand new registrations per month coming out of social services offices—welfare, food stamps, stuff like that—dropped overnight to a couple hundred voter registrations coming out of these offices per month. They just said "hey, forget about that whole registering people who are on foodstamps stuff, that's not working for us."
By the way, all of these things are, in one way, shape, or form, in court today. The gerrymandered redistricting has gone all the way up to the U.S. Supreme Court, and has been kicked back down to the North Carolina Supreme Court. And the Voter Information Verification Act...probably many of you read on Daily Kos about the Winston-Salem trial. That has been NAACP's and the Justice Dept.'s challenge to VIVA. It has gone from U.S. District Court to the U.S. Court of Appeals to the Supreme Court , back down to District Court, and we're now awaiting the judge's decision in that case.
[Audience comment]
Yeah, that evidence of non-compliance with the National Voter Registration Act was submitted into evidence in the Winston-Salem trial, which is a pretty big kick for me.
So, here's the fourth branch of our cutting-edge voter suppression program, that you don't know about, but will in a couple of minutes. North Carolina Republicans have been actively moving the goalposts—they've been moving polling places around like a crazed monkey on crack. They have been cutting numbers of polling places in some counties, increasing numbers in other counties. There has been no systematic analysis of the effect of this. All I've been able to find in any news outlet is, you know: a little local newspaper, say Winston-Salem's, will say: "the number of early voting sites is twelve this year...by the way, it was fifteen last year." That's it. Nobody has taken an overall view.
[Audience comment].
So, I want to talk about this moving the goalposts. My friend, who is a data scientist, suggested to me, "You know, there's all this interesting data out there. We have the names and addresses of all six and a half million voters in North Carolina. We also know the addresses—the geocode locations—of all the polling places in 2014, and 2012. Let's take those 366 Early Voting locations...and I'm focusing here only on early voting sites...let's take the six and a half million voters, figure out where each of those voters' Early Voting location is, and look at the distances between people and polls; how did it change between 2012 and 2014?"
I'll give you the headline right here. I've already told you that under North Carolina Republican governor Pat McCrory, the state board of elections, working with the county boards of elections, decided to reshuffle the polling places in the name of "efficiency" and "cost-savings." And we'll address whether it was either efficient or saved any costs in just a moment, as well.
The headline outcome from our analysis is that in 2014 white voters—71% of the electorate in North Carolina—had to travel an additional 119,000 miles from their homes to their nearest Early Voting locations...which is approximately equivalent to halfway from the Earth to the Moon.
I hear you ask, "how did it affect black voters?" Well, black voters—22% of the electorate—had to travel to the moon and halfway home again, 370,000 miles, in 2014, to get to their nearest Early Voting place.
I'm still having to control myself because, when I first saw these numbers, only a few days ago, I got incredibly pissed. I was sending emails in the middle of the night to Neeta and Denise saying "my wife is out of town and I'm yelling at my dog!" And I'm still incredibly pissed. This is some shameless shit. How did this come to be? How did they do this? We know why they did it, that's obvious, we don't need to talk about why they did it. But how did they do it, and how come you guys didn't catch them doing it?
[Audience comment]
Yeah, well, so much for the cost saving. What it is is the typical passing on of costs—the legitimate costs of government. Is there any more legitimate cost of government than running elections? That's the sine qua non of government, you gotta run elections if you're going to be a democratic government.
So I want to emphasize why distance to poll really matters. It matters for a lot of reasons, the first of which is participatory democracy: the less convenient it is for me to vote, the lower the probability that I will vote. And we have terrible voter turnout in elections nowadays anyway. So doing anything at all that makes it even less convenient to vote is insanity.
Social equity. Rev. Barber was talking last night about how we should make moral arguments; we shouldn't be talking about liberals and conservatives, we should be talking about right and wrong. And the right and wrong of it is that the well-to-do, like many of us in this room, have a much higher degree of mobility and a lot more freedom to say "I'm not coming into work...I'll be a half an hour late, because I'm going to stop off and vote first." Wage slaves in a low-wage job, which a lot of people of color are stuck in today, don't have that opportunity, to tell the manager of McDonald's "I'm going to be a half hour late today because I'm going to vote." So, I really insist that excessive distance-to-poll is a poll tax. It costs you money to go vote, and the more money it costs, the fewer poor people vote.
And it's a civil rights issue. Whether or not you like civil rights as an issue, it's a constitutional protection [....] Any mobility issue disproportionately impacts black voters with respect to white voters. And, by the way, I want to apologize; we've focused this analysis strictly on black voters because that's the big minority chunk in North Carolina, but there are issues for Hispanics, for Native Americans, for a lot of other people of color as well. But I'm just focusing in this talk today on blacks.
And, finally, if you're going to be organizing politically you'd better know...you'd better have geospatial insights with regard to "where are the voters?" and "where are the polls?" and "where are the volunteers who have volunteered to drive voters to the polls, and how are we going to get all these things together?"
So, I think that for all of those reasons distance-to-poll is a really critical parameter.
I'm a scientist. The thing I like about data analytics is that, without it, all you've got is opinions. There's an old joke, which I subscribe to, that opinions are like hemorrhoids—sooner or later every asshole has one. I don't care what your opinion is. I care what the facts are. And data analytics gets you from opinions to facts.
Now, you can't read what's at the bottom of this screen here, but that's one hundred graphs, for all one hundred counties in North Carolina, showing how the median distance-to-poll changed between 2012 and 2014. Don't worry, we're going to show you some highlights you actually can see in a minute; we're not going to go over all one hundred counties right now. What I will say, very briefly, is that we see every different kind of pattern, and I will show you three different patterns from these data. There was no systematic "we're going to go into one hundred counties, we're going to steal polling places from blacks in all one hundred counties." What there was was a rather more efficient and slimy way of doing the same thing.
So here's an example of two of these graphs. What you're looking at here, the Y axis, the vertical axis, is the percentage of voters who live more than ten kilometers from the nearest Early Voting location. I just pulled that number out of my ass, ten kilometers, but I think that's a fair measure of people living on the edge. Ten kilometers is pretty far away from a polling place, particularly if you're poor and lack mobility. And you see a lot of different patterns here. So, in Alexander County, in 2012 there was a real disparity, where 50% of whites lived greater than 10 kilometers from the nearest poll, and only 20% of blacks lived greater than 10 kilometers from the nearest poll. So that was a disparity, advantage blacks. They fixed that! Don't worry...we're on it! It's all OK now. Whereas in Orange County, there was no disparity in 2012, everything was good. They fixed that, too, so that now in 2014 14%—40% more blacks—are more than 10 kilometers from their nearest poll. White voters? We didn't screw around with them at all. So it's good; it's all good.
So we're getting from opinions about "well, that doesn't seem like a very fair place to put a polling place" to facts about what was the impact on the constitutionally protected voting rights of minorities.
I want to summarize all those facts to tell you, first, how they cooked the books: how did they make this happen?
Let me explain briefly: in North Carolina—maybe it's this way everywhere—the county Board of Elections decides where the polls should be for the next election, they submit their proposal to the state Board of Elections, where it is either approved or denied. What the back-channel stuff is about the Governor telling the state Board of Elections what they're gonna do, and the state Board telling the counties what they're going to do, I don't know. But, officially, counties make the decisions, and the state Board either approves or denies them.
So how did they cook the poll books in 2014? It was so simple. It so did not require computers and analytics or anything like that.
There's a lot of different ways we can slice up voters to look at the statistics. The first way I want to slice up voters is, let's look at counties that have a lot of white voters, and counties that have fewer white voters—so let's consider whites first. The top pie chart here is for counties with more than 50,000 white voters in them. Now, that doesn't seem like a lot to you folks from California, but around here that's a fair number of people for a county; this is a rural state. And, nothing outrageous here: for counties with large numbers of white voters, six of those counties saw no change in the number of Early Voting polling places they had in 2014, ten of them lost some sites, and ten of them gained some sites...so, parity overall. There was an average gain of 7% in the number of sites per county for these large counties. I won't talk about the small counties—less than 50,000 whites—right now, because that's really not an issue. So just pay attention to that top [pie chart], the counties that have large urban congregations of white voters.
Now let's do the same slicing, but think about it in terms of black voters; we'll look at the counties that have more than 50,000 black voters. This top graph on the right. There are six such counties in North Carolina. Every fucking one of them lost polling places, such that the average across all six counties was a 31% loss in the number of Early Voting sites in 2014. Every fucking one of them.
So now you see how they did it. The word came down from on high: "Look, fiddle around with the polling places, see what makes sense to you, but be sure that all the major urban centers that black folks live in—Charlotte, Raleigh, Greensboro, Durham, Fayetteville and Winston-Salem—make sure you take polling places away from those cities." And, yeah, you're going to hurt some white voters, but you're going to disproportionately hurt black voters, because that's where the black voters live. [sigh] That's how they did it.
So I want to give you just a couple of case studies so you can get a feeling for these data, because I don't want to walk you through all 100 counties. It's important to see what's going on here.
This first case study is Wake County, North Carolina. It's an important county, it's the home of the state capitol, Raleigh. It's a million people—ten percent of all the voters in North Carolina. It's predominantly white...it's about like the state as a whole: 69% white, 21% black, 10% Latino. The west end of the county...you can't see very well here, but the shading on the maps is, the darker the grey, the higher the percentage of blacks living in that particular area. You can see the western side of the county is white and professional and well-to-do. The eastern side of the county, on the right, is working-class and multicultural, with much higher proportions of blacks. And you can see that, in 2012, in the left-hand picture, there were, I believe that's...sixteen...Early Voting locations in Wake County—not an excessive number for a county with a million people in it. It got revised in 2014 such that there's now nine instead of sixteen Early Voting locations, and as you can see in the eastern half of the county, where all the black voters are, we went from six polling locations serving the eastern black community, to three. Hunh. I'm sure that's completely a coincidence.
So, if you look at the graphs here, the top graph is the median distance-to-poll of each voter. If you don't know what a ‘median' is, pretend I just said ‘the average,' and that's close enough for government work. So, both blacks' and whites' medians increased by about 40%, from three and a half kilometers to five kilometers, in 2014. So, OK, it equally impacted both blacks and whites. And if you look at another way of looking at the data, which I find very interesting, look at the folks at the margins, the people who are furthest away from the polls, the percentage of voters who live more than 10 kilometers from their nearest Early Voting location, that increased for both blacks and whites simultaneously, as well, from just pulling a number of polling places from off of the maps. But, of course, because this is one of the places where many of North Carolina's blacks live, disproportionately it has an effect, because they have lower mobility. Take both blacks and whites further away from the polls, the folks who have the lower mobility and less freedom of scheduling are disproportionately impacted with respect to their convenience of trying to vote.
So, Wake County wasn't broken, but they went ahead and fixed it anyway, making voting more inconvenient for everyone. That's one of the patterns that we see in the data.
Another pattern you can see is in Halifax County, which I drive through all the time going up 95, but I've never set foot in it. It's a rural county with a high poverty rate—32% below the poverty line compared to 17% for North Carolina. Ain't a whole lot of people there. The people there are, are majority black. 53% black, 41% white, 4% Native American, 3% Latino. So, almost two thirds of the people in the county are people of color. There are two little tiny white enclaves, you can see two little white regions, up in the north corners there, one of them is the community of Roanoke Rapids. Interestingly, for a majority people-of-color county, the government is majority-white. The sheriff is white, the district attorney is white, the register of deeds is white, three out of the six county commissioners are white. How does a majority people-of-color county consistently elect a majority white government? One of the answers is how you game the system with respect to whether blacks and other people of color vote or not.
So, in this case again, the top graph is the median distance-to-poll, for blacks and for whites. It's always been unfair for blacks in Halifax County; that's baked into the system, that's how we maintain a white government in a black county. And it didn't get any less unfair, or more unfair, in 2014. Everything was just fine; we didn't touch that county. If you look at the bottom graph, same picture; that's the percent of voters who live more than 10 kilometers from poll. Over a third of black voters live more than 10 kilometers away; less than a quarter of white voters live 10 kilometers away. So, if we go back to the previous [map], you see the yellow circles in these maps mark the Early Voting places. You see one is planted right in the middle of Roanoke Rapids, the white enclave. Gotta have that polling place in the white community! There is no polling place in the black community on the southern side of the [county].
Now, there's a rationale for that: it's a very low density community, these are rural blacks, where would you put it? I don't know...you could put it in a bus and roll it into that area. There's no law that says that a polling place has to be nailed down, is there? But, in all of these maps I'll show you, you'll notice that there's no polling place in any of the black concentrations. But there are plenty of churches to put a polling place in—they'd all be delighted to host one.
OK, so, third and final example, near and dear to my heart, right next door to my county, this is Orange County. Orange is largely suburban-ish/rural. It has a population of 140,000. This is one of the most liberal places in North Carolina, short of Asheville. The population is 77% white, 12% black, 8% Latino, 8% Asian. In the southeastern corner is UNC Chapel Hill, and that helps to make it a liberal enclave in North Carolina. It's highly Democratic—it went for Obama by 42 points in 2012. So, what you can see is the change in 2014, they took a polling place out of God's country in the northwestern part of Orange County, and moved it into God's country in the extreme northeastern part. I live around there, so I can tell you all there are around there is rich folks' horse farms. Why the hell do you put a polling location in an area of rich folks' horse farms? 'Cause it's rich folks' horse farms! And they took the polling place off of the UNC Chapel Hill campus and moved it elsewhere into town just so all the students would have to find time to leave campus and go vote.
So, the result is as you can see. This kind of illustrates how you have to look at a number of different statistics, you can't just calculate medians and go with that. Because if we look at the median, nothing changed much. I want you to notice in the top graph that the scale is very narrow, it only goes from 2.5 kilometers to 2.8 kilometers; so really there's no substantial difference between whites and blacks in Orange County for median distance-to-poll. I'm not going to argue about 0.2 or 0.3 kilometers, that ain't very far—you can throw a ball that far. What I will argue about is the number of voters who live more than 10 kilometers from the poll; the bottom graph. In 2012, again, I've already told you: no problem at all. So we fixed it, and in 2014 we went from 10 to 14% of all black voters, and no significant change for white voters. So 40% more black voters were moved further away from the polls in Orange County. And the students...there's so many different ways you can analyze these data. You can look at it by age, by income as well.
So, what I want to emphasize here is that this kind of data analysis is not just good for giving you something to wring your hands about, or get pissed off about. It suggests solutions to problems. The kind of solution I would like to propose is to take this approach to launch what, for lack of a better name, I'll call the Fair Places Project. Why can't we put together an organization that uses geospatial data science to document disparities such as I've shown you, and then goes to the relevant county boards of elections and says, "Excuse us...you have a problem in your county. You have an enormous racial disparity with respect to distance-to-poll. And, gosh, that's unconstitutional. And we'd like to help you with that." This is point three here. We give them a carrot. We say: "We can give you advice on where you can drop one or two more polling places in your county in order to have no racial disparity. You won't have to worry about doing the computing; we'll do the computing for you. We'll show you all your options for making that disparity go away. We'd love to help you." And if—and when—they tell you to go fish, then you whip out the stick, and you sue their asses off.
Additionally, an organization like this could provide the Democratic Party, candidates, and activists with geospatial voter insight. Where has [voting] gotten hard? Which counties, which districts, which precincts has it gotten difficult for voters to get to the polls in, where we should concentrate our get-out-the-vote efforts, get volunteer drivers to take people to the polls.
That's all I've got. Thank you very much.