The American working class is badly misunderstood, criminally underrepresented, and, like any voting bloc, shot through with division. But if there’s one thing workers can agree on, it’s their collective opposition to the guy who controls their paychecks, their time, and even their bodies.

Given what we know about his experience at Bain Capital—the Wall Street firm that profited from outsourcing jobs overseas, leaving behind scorched townships in battleground states like Ohio—that guy looks increasingly like Mitt Romney.

Let’s consider, for illustration, two campaign ads. Both are about GST Steel in Kansas City, Missouri. On the one hand, is a story told by craggy-faced steelworkers about slick-back paper pushers sucking the life out of the mill.

They load it up with debt, force it into bankruptcy, and then renege on a promise to honor pensions and health care. Bain walks away with $9 million before the ink is dry.

The other story is about a heroic effort to save a business and the dream of a mill that could sustain itself. “There’s this vampire story that Bain comes in…and sucks the blood out,” a man says. “It was really entirely the opposite… We went looking for a blood donor.”

I don’t suggest either is right or wrong. I merely note two things. The first ad is about bad consequences: Bain closed the mill, made a buck. The second is about good intentions: Bain tried its best.

The other thing? That the second ad was made by the Romney campaign to defend Bain’s deal with GST, and that it features just one person, B.C. Huselton, GST’s former vice president.

In other words, the boss.

Among the many stories about the white working class in the U.S. is this dominant thread: In the 1970s, working-class voters defected from the Democratic Party in reaction to the civil rights movement, merging with the conservative backlash that evolved over decades into a working-class movement that undermined working-class interests.

Thomas Frank masterfully explored this theme in his 2004 best seller, What’s the Matter with Kansas?, explaining how George W. Bush beat John Kerry by invoking the triple threat of “gays, guns, and God.”

Four years later, then–Senator Barack Obama followed suit, explaining (in comments not intended for the press or public) why the white working class votes against its interests: “People end up, you know, voting on issues like guns…like gay marriage. And they take refuge in their faith…and things they can count on [because] they don’t believe they can count on Washington.”

Is this true? Is the white working class more conservative, more concerned about cultural issues, and therefore more Republican? It depends on what you mean by “working class.”

Class is usually defined in terms of education, wealth, or geography. But according to Michael Zweig, an economics professor at the State University of New York at Stony Brook, these categories can be misleading or worse. The best frame for understanding class, he says, is power:

“The working class [has] relatively little power at work—white-collar bank tellers, call-center workers, and cashiers; blue-collar machinists, construction workers, and assembly-line workers; pink-collar secretaries, nurses, and home health care workers skilled and unskilled, men and women of all races, nationalities, and sexual preferences,” he wrote in the Monthly Review in 2006.

Zweig says two forms of power fundamentally separate the working class from other classes (he identifies four in all). One is the power to control the terms of your work: where, when, for how long, and for how much. The other is the power to work without constant supervision.

When the boss is telling you what to do, when to do it, where to be, what to wear, how to talk, and even what to think all day every day, you may learn to live with it, but you don’t have to like it.

Or the boss.

Power, of course, correlates strongly with income, and income correlates with voting patterns, writes Dorothy Sue Cobble in Dissent. The bottom 45 percent of income earners, white and otherwise, has voted Democratic since the 1930s, Cobble says. Only with incomes around $50,000 a year does party affiliation begin to diverge, and the higher you are on the income ladder, the likelier you are to vote Republican. But what about the white working class?

According to political scientist Larry Bartels of Princeton, the bottom third of white income earners has grown more reliably Democratic over 30 years. In fact, if you separate the South from the rest of the country, he says, you find that Republican gains have come from the middle and upper-middle classes, while Democratic losses “among poor whites over the past half century [are] entirely attributable to the demise of the Solid South.” (Italics his.)

Moreover, Bartels says, economics trumps moral issues, not—as we are so often told—the other way around. Indeed, the farther down the income hierarchy you go, the truer this becomes. The higher you go, conversely, the more moral issues trump economics.

And income holds sway even in states where moral issues, like abortion, tend to dominate, according to research headed by Columbia University’s Andrew Gelman. They found that rich people in poor states like Kansas are much more likely to vote Republican than their counterparts in rich “blue America” states like Connecticut, where the wealthy vote for reasons other than economics.

As Gelman and his colleagues put it in their study: “The Mississippi electorate is more Republican than that of Connecticut; so much so that the richest segment of Connecticutians is only barely more likely to vote Republican than the poorest Mississippians. In poor states, rich people are very different from poor people in their political preferences. But in rich states, they are not.”

So: Those with “relatively little power at work,” or the working classes, vote overwhelmingly Democratic, while those with relatively more power at work, the middle class and/or the bosses, don’t.

Power tells us something about wealth, which tells us something about voting patterns, but that’s not all: Power can tell us about a candidate’s likability. In fact, power can tell us about the opposite.

Likability is that nebulous variable of presidential elections that no one can quantify yet everyone knows is crucial. With the exception of Richard Nixon, no candidate in the modern era has won without high likability ratings. And favorability, a catchall trait among pollsters that includes likability, has been an accurate predictor in the last five presidential elections.

According to Gallup’s latest poll, Obama has 54 percent while Romney has 46 percent. This worried Romney in May when Gallup released a poll showing that 60 percent like Obama while only 31 percent like Romney, a 29-point margin. It was better in June, but not much: 64 percent for Romney, 81 percent for Obama.

Romney, of course, has a chance of winning white working-class voters, especially those susceptible to populist conspiracism (think birthers). But the more we learn about his former private equity firm and its history of maximizing profits while minimizing jobs, the more his likability will erode.

The GST Steel ad was the first to tap into the uncharted depths of working-class resentment and speak to those decimated by the likes of Romney, who are likely to vote Democratic. But even the rare outlier still must ask if he likes the guy who looks like his boss. And nobody does.

John Stoehr is a lecturer in political science at Yale and a frequent contributor to The American Prospect, The New Statesman, and The New York Daily News. He last wrote for the Spectator about the book Land of Promise: An Economic History of The United States.

Also in this issue: Bain Capital executives’ real opinion of their employees.