(This 7 page document is in PDF so I will copy the first few pages.)
January 19, 2005
VOTING WHILE BLACK (AND BROWN AND RED)
How the color of your skin still determines whether or not your vote countsby Jordan Green
If you are African American, Latino or Native American and voted in the 2004 presidential election that allowed George W. Bush to keep the White House, you did so with a higher likelihood that your vote wouldn’t be counted than if you were white.
That analysis comes from a review of residual voting statistics in heavily minority counties in Ohio, Florida and New Mexico – that is, the number of people who showed up to vote, but didn’t have a preference for president counted.
But before you assume this means there’s a Jim Crow-style system of electoral exclusion designed to systematically suppress the minority vote, stop. Before you assume this means the election was stolen from John Kerry, stop.
Almost any rigorous examination of the methods of counting votes in Election 2004 by race, party clout and political subdivision suggests that Bush legitimately won reelection. But the absurdity of some of the conspiracy theories thriving in the progressive wing of the Democratic Party shouldn’t distract us from the real flaws in an electoral system that tends to dilute the power of the minority vote. We should be clear that while we have made progress towards making sure every vote counts, the United States still has a long way to go to make sure minority votes count as much as white votes.
To varying degrees, minority voters in Ohio, Florida and New Mexico were less likely to have their votes for president count than their white counterparts in Election 2004. The reasons were complex. They probably include technology, as well as unequal funding for poll workers and voting machines. To some extent, as in New Mexico, intentional disregard for the presidential slate of the ballot also plays a role in diluting minority voting strength.In Ohio, voting technology seems to have been a strong determinant of how well minority votes were counted. Two heavily Democratic counties in northeast Ohio, Cleveland’s Cuyahoga County and Akron’s Summit County, had abnormally high residual voting rates: 2.0 ercent and 1.9 percent, respectively. Both used antiquated punch-card voting machines. In two other populous Democratic counties in northern Ohio, Mahoning and Lucas, votes for president were much more likely to count. Mahoning County, home of Youngstown, which uses ES&S’ touch-screenvoting machines, had a residual voting rate of 1.1 percent. Lucas County, which encompasses Toledo and uses mostly optical scan machines, had a residual voting rate of 0.6 percent.
Some of the residual vote can be chalked up to voters intentionally choosing to vote for none of the presidential candidates. About 30 percent of votes not cast for a presidential candidate can be attributed to intentional neglect, according to a 2001 report by the CalTech/MIT Voting Technology Project report. Their estimate is based on exit polling data from previous elections.
Most of the residual vote, however, results from voters who believe they’ve expressed their democratic choice who accidentally vote for two candidates, or accidentally vote for no candidate because a punch card machine doesn’t fully perforate the card, an optical scan machine doesn’t pick up a mark, or a touch-screen voting machine skips past the presidential slate page before the voter can make a choice. Touch-screen machines at least have the advantage of not allowing voters to choose two candidates, thus eliminating the opportunity for over-voting.
If there is one place in Ohio where the structural racism of spoiled ballots might wrest the presidency from a Democratic candidate, it is Cuyahoga County. Heavily unionized, battered by recent industrial job loss, and 27 percent black, the northeastern county is the largest reserve of Democratic votes in the state. Combining majority black Cleveland with its vanilla suburbs, the county delivered 33,262 votes to Kerry. As the largest county in the state and the 19th largest in the nation, Cuyahoga also produced the most residual votes of any county in Ohio: 12,933. Assuming the Voting Technology Project rule is correct in its estimation that 30 percent of the residual vote can be attributed to intentional neglect of the presidential slate of the ticket, and considering the share of votes that went to each candidate, a conservative estimate would restore 6,337 votes for Kerry in Cuyahoga County.
Factor in an estimated 2,640 votes lost by Bush in the undercount, and Kerry might well have lost 3,697 votes.Notwithstanding the fact that Bush appears to have legitimately won the election, an analysis of the residual vote in Cuyahoga County does indicate that ballot spoilage disproportionately disenfranchised African-American voters. In Cuyahoga County cities, ownships and villages where whites constitute a majority, the average residual voting rate was 1.4 percent. But take majority black Cleveland and a half dozen majority black municipalities like East Cleveland and Warrensville Heights as a whole and the residual voting rate jumps to 2.7 percent. Voters in majority-black Cleveland and other
majority-black municipalities in Cuyahoga County were almost twice as likely to lose their vote for president as voters in the county’s predominantly white suburbs.
The voting patterns and ballot spoilage rates in Cuyahoga County’s black communities suggest that in a much closer election,
minority vote suppression – especially with the kind of vote challenger campaigns the Ohio GOP threatened as late as the day before the election – could have the effect of fouling the result. Small though it was, the county’s net vote gain for Kerry was concentrated in the predominantly African-American urban core, not in the white suburbs where the vote was more evenly split between Kerry and Bush, and where the residual voting rate was smaller.
In communities such as Bedford Heights and Warrensville Heights, both upwards of 90 percent black, voters favored Kerry to Bush, 95-5.
Restoring spoiled ballots for those two communities, Cleveland and four other majority-black municipalities would likely give Kerry a net gain of 2,387 votes. In contrast, the county’s majority white communities – ranging from 60 percent white Shaker Heights where the Democratic candidate won 77 percent of the vote, to 99 percent white Hunting Valley which yielded only 27 percent of the vote to the Democrat – would net Kerry would only 769 votes.
Of course, because of the right to a secret ballot, it’s usually impossible to go back and ask particular voters to whom they really intended to give their vote. The one exception is when voters check off the candidate of their choice and also write the candidate’s name on the ballot as a write-in candidate. An optical scanning machine would automatically reject such a vote, but any election worker of reasonable intelligence who examined the ballot would recognize the voter made a clear choice.
At least one other Democratic area of the state also appears to have taken a hit in high residual voting rates, potentially cutting into Kerry’s column. The Dayton Daily News reported in November that the residual voting rate in the 231 precincts that supported Kerry in Dayton’s Montgomery County averaged 2.8 percent, but only 1.6 percent in the 354 precincts where Bush won majorities. Overall, the residual voting rate in Montgomery County, which used punch-card
ballots, stood at 2.0 percent. In two precincts where majorities voted for Bush, 27 percent of ballots cast did not register a vote for president, although
undercounts tended to hit Democratic voting areas to a greater extent than Republican strongholds overall.Comparisons between counties in Ohio suggest the inferior technology of punch-card voting machines disenfranchises voters without regard to race. Overall, rural Appalachian Republican counties appear to have been affected just as badly if not worse than the northeastern Democratic bases where union and African-American support for Kerry was strong. In the 10 counties that Kerry carried by the largest margins, 1.4 percent of all ballots cast did not register a vote for president. In the top 10 Ohio Republican counties, the residual voting rate was even higher, at 1.9 percent.
continued in PDF format here
http://www.southernstudies.org/reports/VotingWhileBlackJan05.pdf