How Philadelphia’s Fuzzy Racial Math Could Tip the Mayoral Election

Photographs by Jeff Fusco

Eight months into Michael Nutter’s first year as mayor, the New York Times Magazine published an 8,200-word story headlined “Is Obama the End of Black Politics?”

It was a deliberately provocative question, and one that resonated in Philadelphia. Nutter, who featured prominently in the story, had won office a year before with an unprecedented multiracial coalition. Barack Obama had just delivered his remarkable speech on race — an address that seemed to convince many white voters to support him — at the Philadelphia Constitution Center.

At the time, it seemed very much as though Philadelphia’s long tradition of racial politics — that is, the tendency of whites and blacks alike to vote mostly for candidates of the same race — was winding down.

Well, not so fast.

The role of race in local politics — which, let’s face it, never remotely went away — has moved to the center ring of Philadelphia’s latest mayoral election.

Two weeks ago, a group of influential black politicians from Northwest Philadelphia, including Dwight Evans and Marian Tasco, endorsed Jim Kenney, an Irish dude from South Philly who spent most of his adult life as a mummer, was schooled by Vince Fumo, and is now backed by Johnny Doc.

This week came the backlash. George Burrell, a longtime adviser to John Street and a mayoral candidate himself back in 1991, proclaimed at a press conference that the city’s pioneering black political leaders were “rolling over in their graves” at a move that could “turn over the mayor’s office, voluntarily — voluntarily — to a community other than our own.”

Prominent black business leader A. Bruce Crawley was similarly enraged. He described the Dwight Evans endorsement of Kenney to me as “a bit like betrayal” and said: “It’s common wisdom that, with the number of black registered voters in this city, if black voters function in a monolithic way, the black candidate, or the candidate that is supportive of black issues, will win.”

How widespread is the backlash against Kenney’s bid to win significant black support? It’s hard to say. It’s equally difficult to predict how much Kenney will benefit from the endorsement of a handful of black elected officials. Voters tend to make up their own minds in mayoral contests, and endorsements have questionable value.

Nonetheless, a lot of political observers think that the 2015 election, like 2007’s, will feature a higher-than-typical amount of cross-racial voting. They cite Lynne Abraham’s name recognition and some internal polling she’s done that she says shows her with significant black support. In Kenney’s case, they point not just to the Northwest coalition’s endorsement, but his union support as well.

Unlike in the Nutter and Obama contests, however, few political pros expect many white voters to support Williams, who is the leading black candidate (Doug Oliver could run well in some white neighborhoods, but he is barely registering in polls so far).

All of which explains the tension among black political leaders over the stirrings of support for Kenney. Crawley is correct that, if black voters act like a monolith, and, just as crucially, if they turn out to vote in solid numbers, Williams will win. If the black vote fractures, Williams will lose.

In other words, the black voters who cross the color line when casting their ballots could decide this election. The question is, how many of them are there likely to be?

History’s Answer: Not Many

At Philadelphia magazine’s request, Kevin Gillen, a senior research fellow at Drexel University, analyzed racial voting patterns in five competitive, high-profile Democratic primary elections that featured a multiracial field between 1991 and 2010 — three mayoral races, the 2008 presidential primary, and the 2010 gubernatorial contest. His research confirms that there is an extraordinarily powerful association between racial identity and voting in Philadelphia, one that has persisted more tenaciously than many assume.

Gillen performed something called a regression analysis, which is essentially a way to empirically measure relationships between variables like voting and race (you can read more about his analysis, and its inherent limitations, at the end of this story). He found extremely close relationships between votes cast and the racial makeup of individual wards in Philadelphia. That close association existed in both majority-black and majority-white wards.

Of the five races, it was the 1991 mayoral primary featuring Ed Rendell, Lucien Blackwell and George Burrell that showed the most extreme racial polarization in voting. That polarization remained high, but declined measurably, in the 1999 Democratic mayoral primary featuring John Street, Dwight Evans, Marty Weinberg, and Happy Fernandez.

Eight years later, the mayoral field of Nutter, Chaka Fattah, Evans, Bob Brady and Tom Knox scrambled racial voting patterns in a historic way. Although race was still a strong predictor of voting, the race-to-vote correlation was 74 percent lower for white voters and 45 percent lower for black voters. Those results were a testament to the cross-racial appeal not just of Nutter, but of Brady and Knox as well, both of whom won significant votes in a number of predominantly black wards.

The 2007 campaign proved conclusively that certain candidates are capable of defying the city’s “racial math.” But do any of the candidates in the 2015 field have the crossover appeal of Nutter? That’s not at all clear.

It’s important not to overstate the lessons of 2007. Even then, most black and white voters alike cast ballots for candidates of the same race. And the historic norm of highly racially polarized voting began to return the very next year, in the Obama-Clinton matchup (though Obama garnered a significant number of votes from largely white wards, Clinton did not perform well in predominantly black wards).

By 2010, the old racial voting patterns were back at nearly full strength. Anthony Williams, then running for governor, received staunch support, albeit in a low-turnout election, from black wards. But Williams was unable to carry a single majority white ward, in spite of the fact that he was the only Philadelphian in a four-candidate field.

Race and 2015 Election

This is why nobody is writing headlines suggesting Williams is the new leader of white Philadelphia, as the Daily News did two weeks ago after Kenney received the Evans endorsement: “Is Kenney the Future Voice of Black Philadelphia?” Over at Newsworks, Dave Davies called the endorsement “historic.” Here at Citified we surmised that it could be “the most critical endorsement” of the race. And maybe it will be. One of the super PACs supporting Kenney has a new ad touting the endorsements that is clearly aimed at black voters. The message? Sure, he’s white, but it’s OK to vote for this guy.

And yet, it’s actually not all that unusual for prominent black pols to endorse white candidates for mayor. Councilwoman Jannie L. Blackwell backed Tom Knox in 2007. John F. White Jr. — who ran in the Democratic mayoral primary in 1999 — endorsed Sam Katz against John Street that same year (Katz, remember, was a Republican to boot). And, ironically, Williams’s father, the trailblazing Hardy Williams, endorsed Ed Rendell in 1991.

So how best to understand the backlash over the Northwest coalition’s endorsement of Kenney? It seems likely that it’s either one of two motivations: 1) Simple politics. Burrell, for instance, is backing Williams. And by shaming Evans and others for endorsing Kenney, he may make other black politicians think twice before doing the same. 2) Genuine worry that, this time, the black vote could fracture in a meaningful and perhaps lasting way. Or perhaps it’s a mix of both.

In other quarters of the black Philadelphia political class — and not just in the Northwest — the thinking appears to be a bit different. Philadelphia has had three black mayors, each of whom served two full terms. That fact, plus the other powerful positions held by African-Americans in city government, appears to have taken some of the urgency out of the inclination to support the leading black candidate for mayor; at least in some quarters. Mayor Nutter seems to have played a role here as well. Right or wrong, there are plenty of black political leaders in the city who doubt his commitment to the issues they consider most critical to black Philadelphia. Electing a black mayor isn’t an end unto itself, in other words. And there are black political leaders with profound policy differences with Williams over education policies (though Dwight Evans definitely hasn’t been one of them). Will they overlook those differences to ensure a black mayor succeeds Nutter?

Crawley, who donated $2,000 to Williams’s campaign last year, said it’s not about race per se, but about supporting a candidate who’s committed to policies that will help the black community. At a black political summit last weekend, Crawley said, the talk was about policy: “Nobody talked about whether the candidate was black, red, yellow. I think that reflects a growing sophistication.”

At core, though, this is the reason that racial voting patterns are as strong as they are, in Philadelphia and elsewhere and among both blacks and whites. Former mayor Wilson Goode explained it like this: “Black people, like white people, will vote for their interests. I don’t think African-Americans will blindly vote for a black candidate without regard to whether they’re qualified. I don’t think there’s any possibility that an unqualified black candidate would defeat a very, very qualified white candidate in Philadelphia.

“But with everything else being equal, in terms of qualifications, I think black voters for the most part would make the assumption that the black candidate is the one that best understands their interests.”

Those interests were in stark relief in 1983, when Goode became Philadelphia’s first African-American mayor. The Democratic primary featured Goode and Frank Rizzo, the former police commissioner and two-term mayor who was a walking personification of big-city racial strife. The black vote very nearly was a monolith in 1983, and it was just big enough to narrowly defeat Rizzo.

Neither the black vote nor the white vote are monoliths any longer, even in the most racially homogenous of wards. Michael Hagen, a professor of political science at Temple University, looked at Gillen’s data and zoomed in on those wards where the populations were more than 80 percent black or white — even in those areas, the trendlines show a small decline in the association between race and voting results.

“My guess, and this is only a guess,” Goode said, “is that there will be in the neighborhood of 20 percent of black voters that could cross racial lines in this primary. That would be high. I don’t think it would be any more than that.”

Kenney and Abraham are counting on more. Williams hopes for less. That swing, however big it ends up being, very well could determine the identity of Philadelphia’s next mayor.

Kevin Gillen’s regression analysis uses voting data from the City Commissioners office and demographic data from the decennial census closest in time to each analyzed election. The census data includes only adults 18 years or over.

There are certain limitations to Gillen’s analysis. The census data represents all residents of a ward, not registered voters. Additionally, there may be turnout differences between black and white voters that are not captured in this analysis, because voter registration records do not track race. There are other social factors that may influence voting patterns that are also not accounted for in this analysis, such as income and education level.

Nonetheless, Gillen is confident that the high correlation between ward demographics and voting results strongly suggests a powerful link between racial identity and voting patterns.

Special thanks to Casey Thomas for his work putting together the interactive map.

Later this week on Citified, look for a story analyzing the Latino vote and its potential impact on the 2015 mayoral contest.