In 2012, one of the running and possibly distasteful jokes among political reporters went something like this: Republican National Convention organizers managed to allot time, space and national attention to what seemed like every black, Latino and Asian American GOP elected official they could find — even as there were very few among the delegates present.

If that was in fact the idea, it didn't seem to do much good. That November, Mitt Romney lost the election and all of the fast-growing, non-white parts of the electorate by expansive and in some cases long-unseen margins. The results set off a kind of semi-public wave of party introspection that went well beyond the typical failed-election postmortem. One of the resulting report's big recommendations boiled down to this: Cultivate and field more candidates of color so that non-white voters will consider casting more votes for Republicans.

It makes sense in theory, but perhaps less so in the real world.

For black voters, a candidate's race can make something of a difference, but party and policy positions matter much more, argues a new University of Cincinnati study presented at an American Political Science Association conference this month.

How did David Niven, the political scientist behind the study, reach such a conclusion? A real-world, real-time election experiment.

In 2014, a number of high- and low-profile races were slated for ballots in Ohio. In Franklin County, which includes Cincinnati, black Republican candidates were vying for some of those downballot offices. These are races that typically do not receive much local or national media coverage. Niven saw this as an opportunity to test what, if any, influence very basic factual information had in 28 precincts where all registered voters are black.

Niven divided the voting districts into three groups. He sent the following information to these groups about a black Republican incumbent candidate seeking the county auditor slot and a black Republican vying for an open judgeship.

  • Group No. 1: No information
  • Group No. 2: Fliers featuring candidate photos released by the respective campaigns, candidate names and the offices sought.
  • Group No. 3: Fliers with candidate photos, names, offices sought and the candidate's party affiliation.

Niven found that in precincts where black voters received that last kind of flier — an "Endorsed by the Republican Party" leaflet — African American GOP candidates fared far worse than they did in areas where voters received no information or fliers with just the candidate's photo, name and office sought.

Niven thinks the results were especially instructive in the judicial race, because candidates running for the court of common pleas are listed on the ballot without party affiliation. In that race, black voters who received a flier that did not list party affiliation (Group 2) were nearly three times as likely to vote for a black Republican on Election Day. And voters who received a flier that did list party label (Group 3) or no flier at all (Group 1) were actually considerably less likely to support black Republicans than they were the white Republican at the top of the ticket, Ohio Gov. John Kasich.

Niven found that in precincts that received fliers with photos (Group 2), black Republicans gained 15 percent more black votes than Kasich. In other words, voters who probably knew the candidate's race but not the candidate's party affiliation (Group 2) were more likely to support black Republicans than those who had no information (Group 1). At the same time, in precincts where voters received no information (Group 1) as well as those where voters received fliers with photos and party affiliation (Group 3), black Republicans finished well behind Kasich.

In other words: Race and party information produced no advantage for black Republicans among black voters. For black voters, race matters, but party matters much more.

Why?

Before we explain, a few key notes might be useful here.

First, in 2014 Kasich, did very well — for a Republican — among black voters. He won 26 percent of Ohio's black vote, but that's surely in part because he benefited from a series of scandals, questions and controversies that dogged his Democratic opponent, Ed FitzGerald.

Secondly, black voters are more likely to support black candidates just as white voters are more likely to back white ones. That's an undeniable truth borne out in more than a few elections and Ivory Tower examinations of election outcomes. Niven's study was small, focused on 28 all-black precincts in Ohio. So the results can't be applied with certainty to other groups — Latinos, whites, Asians, LGBT citizens, etc. Niven hopes to identify elections in which he can test those possibilities in the near future.

Now back to those reasons.

Niven has some ideas. Black Republicans seeking elective office rarely embrace or publicly express a broader range of ideas, including those that would put them closer to the center or center-left of the party, Niven said. So the beliefs that many African American voters may have about Republican candidates and the party go unchallenged or are often affirmed during elections.

"The kind of folks, the African Americans who tend to run as Republicans," Niven said, "they aren't usually from the John McCain or Colin Powell wings of the party. They typically come from the far right. They are the Tim Scotts, the Allen Wests, the Mia Loves." Scott is a senator from South Carolina, West is a former congressman from Florida and Love is a congresswoman from Utah.

What does that mean?

Here's a pretty clear example: In the weeks after Love won her seat, she became the first black Republican woman in Congress, and received all the media attention you'd expect to come with that. The spotlight stayed a bit longer when she publicly backed new House Majority Whip Steve Scalise (R-La.), despite reports that he gave a 2002 speech to a white supremacist group founded by former Klu Klux Klan leader David Duke. (Scalise and the event's organizers told reporters that Scalise was unaware of the group's ties when he appeared.)

"She didn't simply say nothing or say, 'No, I'm not going to try to knock him out of leadership,' " Niven said. "Mia Love said, 'I have no problem with it.' For black voters, it’s one thing to not pursue the issues that the NAACP cares most about. But when you are not particularly concerned that someone is close to David Duke in the leadership of your party, there is something peculiar about that."

Now, to be clear, some people aren't sold on Niven's conclusions.

In fact, one of the black Republicans who ran in one of those races in Franklin County, Ohio, and won reelection had some pretty strong things to say after news of the study went public. In a statement issued this week, Franklin County Auditor Clarence Mingo said:

Niven says some African-American candidates are more acceptable than others, he creates a distinction not unlike the old ‘house negro’ versus ‘field negro’ or ‘smart negro’ versus ‘unintelligent negro.' This is stereotyping of the worse possible sort and is totally inappropriate, inaccurate and irresponsible.

Mingo's statement also quantified his performance among African American voters. He outperformed other Republicans, according to his statement. In 2014, he earned an average of 28 percent of the black vote in precincts where 50 percent of the voters were African American.

Whatever the results, we could soon see Niven's study play out on the national stage.

Sen. Ted Cruz (R-Tex.) is half-Cuban and Ben Carson is black. Louisiana Gov. Bobby Jindal (R) is Indian American. All three men can pretty definitively be described as allegiant to very conservative Republican ideas. Sen. Marco Rubio (Fla.), a Cuban American, also tends to hew pretty closely to the party's orthodoxy on just about every major issue in the presidential race — save perhaps for his push for comprehensive immigration reform (which he has notably walked back). And Cruz, Jindal and Rubio all have policy records that lean more right than center-right.

But, Niven noted, Rubio has stayed away from much of the conservative conversation about mass deportation and border walls this summer. That's probably a good start.

As for the other candidates — or even Rubio, for that matter — there's little evidence that they are bringing minority voters into the GOP. In fact, about the only candidate who appears to have that kind of crossover appeal is former Florida governor Jeb Bush, who is white but has a biracial family with his Mexican American wife.

"For the upcoming election," Niven said, "the bottom line is unless the Republicans — the party and its candidates — actually speak to the concerns of the people whose votes they are targeting, they are unlikely to win those votes."