Abstract
If men can respond to a threat to their masculinity by refusing to do the dishes, doing push-ups, or going to a more fundamentalist church, why not respond by espousing certain political views or favoring certain candidates?
Keywords
Masculinity is a brittle thing: strong, but intolerant of flexibility, and so when men find their traditional gender roles threatened by something—loss of economic dominance and fear of being perceived as homosexual seem to be common triggers—they have been shown to engage in otherwise inexplicable behavior. When men find that they’re no longer fulfilling one of the major aspects of their gender identity, because they earn less than their wives, for instance, they do less housework, they go to traditionalist churches more, they do more push-ups, and they express stronger anti-gay views. Now, with Donald Trump in the White House, I can add another: men who perceive that their masculinity is under threat change their political behavior.
Men earn more than women for the same work, get promoted faster, and even pay less for goods that require negotiation, like cars. But this hasn’t stopped men from believing that their gender is under siege.
In some sense, it seems silly to think of men’s economic dominance as threatened in the United States. After all, 492 of Fortune 500 CEOs are men. Men earn more than women for the same work, get promoted faster, and even pay less for goods that require negotiation, like cars. But this hasn’t stopped men from believing that their gender (along with markers of its “natural superiority”) is under siege: in the 2016 American National Election Pilot Study, fully 40% of Republican men said that they faced “a great deal” or “a lot” of discrimination because of their gender. Republican men, especially, have come to believe that they’re being victimized. This view is leading them to fight back, and politics is one of the most potent tools in their arsenal. Using survey experiments and panel data tracking men over several years, we can see how economic and social threat are pushing American men toward conservative policies unrelated to gender and very much against Hillary Clinton.
It helps to understand why masculinity seems to be simultaneously so strong, and yet so prone to threat. Masculinity, unlike femininity, is earned, often through trials or rituals, and can be lost if a man fails to live up to the demands of his gender. Women’s gender identity, on the other hand, is more often linked to biological than behavioral factors. This means that men’s gender identities are potentially always under threat and that men must guard against the loss of masculinity at all times. So, when men are failing to live up to some aspect of the traditional masculine gender norms—what’s been termed “hegemonic masculinity”—there’s enormous pressure to respond in some way, often by doubling down on some other aspect of masculinity. A classic example comes from an experiment reported by Jennifer Bosson, Jennifer Prewitt-Freilino and Jenell Taylor in 2005. In it, men were asked to carry out a task in which they had to braid strings together. When the task was described as “rope-reinforcing,” the men who had completed it faced no threat to their masculinity; but when it was described as “hair braiding,” the men in the study became more prone to verbally assert their heterosexuality.
Gender-Priming Political Match-Ups
Source: 2016 telephone survey of voters in New Jersey, author’s own. Data available upon request.
Men have been repeatedly shown to engage in these “compensatory manhood acts” in the laboratory, but men in democracies have another, potent way in which to assert their masculinity in the face of threat: politics. If men can respond to a threat to their masculinity by refusing to do the dishes, doing push-ups, or going to a more fundamentalist church, why not respond by espousing certain political views or favoring certain candidates?
To study how men might use politics to respond to a threat to their masculinity, an experiment was embedded in an otherwise normal 2016 telephone survey of voters in New Jersey. In the course of the survey, respondents were asked who they would favor in various match-ups for the then-upcoming Presidential election, but not before half of the respondents were randomly assigned to get a question prompting them to think about gender norms. Specifically, the question said that there are an increasing number of households in which women earn more money than men, before asking about the respondent’s household. The expectation was that many men might lie in response to this question, but their answers weren’t the point: the point is to prime them to think about threats to their gender identity. Men who weren’t primed to think about gender norms favored Hillary Clinton in a match-up against Donald Trump by a 16-point margin—nothing unusual in a Democratic-leaning state (see above). But those men who were primed favored Trump by 8 points in the same match-up. Nor was this purely a case about money or gender pushing men toward the Democrats: the priming made no difference in the margin of support for Bernie Sanders over Donald Trump. In essence, the threat of losing the traditional norm of men as breadwinners led men to abandon support for the first major-party female candidate in American history and come out in support of her opponent.
Letta Page, Flickr CC
The size of these results was striking enough that a partial replication was attempted in a national telephone survey the following month. In it, respondents were asked a series of questions about women’s status in society, leading to a more diffuse threat, rather than the very specific threat that would come from making less money than their wives. Again, half of the respondents were given such priming questions before answering Presidential match-ups and half afterwards. Among those men who believed society is biased against men—22% of respondents said that the media treats women less harshly than it treats men—the priming reduced support for Clinton by 6 points in a match-up with Trump, with no significant effect on support for Sanders.
Trump’s campaign slogan suggested a threatened “greatness” and promised its restoration.
Gage Skidmore, Flickr CC
Generally supporting the same findings as the previous experiment, this survey revealed an interesting twist that wasn’t seen in the initial experiment. While men who see the media, and, by extension, society, as being biased toward women were less likely to support Clinton when primed to think about gender norms, there were also men who said that they saw the media as being biased against women. Those men, when primed to think about gender norms, were more likely to say that they supported Clinton in a match-up with Trump—and also more likely to say that they supported Sanders, though by a smaller margin. This suggests that threat to traditional gender norms aren’t impacting all men’s political views in the same way—it is leading some men to be generally more liberal.
Together, these experiments indicate that threats to gender norms, arising from both the potential individual threat of losing relative income and overall societal threat from institutions being biased against men and in favor of women, drove men away from Hillary Clinton and toward Trump, in the 2016 election. But survey experiments, compelling as they may be, aren’t enough to establish that men are using politics to compensate for gendered status threat. Moreover, there’s the hint of polarization, the idea that different men may react differently to challenges to gender norms. To dig into these issues, I used the 2006-2008-2010 General Social Survey (GSS) panel study. In it, respondents were contacted up to three times, at two-year intervals, over a period that included the height of the great recession. Purely by accident, this turns out to be an ideal period in which to measure the impact of threats to male economic privilege, as the initial wave of the recession disproportionately hit men, rather than women. The male unemployment rate rose to more than two points above the women’s unemployment rate for more than a year; that 2% difference represents millions of households in which men lost their jobs, while the women did not. Indeed, more than half of the men in the panel reported having a lower income, relative to their wives, than in the previous wave of the panel. And 20% of men lost more than 25% of the relative income in the household, dropping, for instance, from contributing 60% of the total household income to 35%.
If men can respond to a threat to their masculinity by refusing to do the dishes, doing push-ups, or going to a more fundamentalist church, why not respond by espousing certain political views or favoring certain candidates?
Clinton’s campaign emphasized the history-making moment of a major-party female candidate.
Tim Evanson, Flickr CC
Just as income was measured at up to three time intervals, so, too, were a selection of political and social views. The fact that the same individuals were asked at multiple points in time means that the analysis can track how changes in their relative income correspond with changes in their political views over the same time. This eliminates selection bias, a major problem with most studies looking at relative income. For instance, it’s entirely possible that men who choose to marry women who make more money than they do or have the capacity to do so are different from men who choose to marry women who don’t make more money than they do. Indeed, at least one large-scale analysis has found that loss of relative income only matters to men who earned more money than their wives at the start of the relationship. As such, any analysis that looks at differences between the men who make more and men who make less than their wives might be driven by pre-existing differences between the groups. By looking at the same men at different points in time, it’s possible to be certain that the differences are driven by changes in the individual, rather than the difference between two groups.
An increasingly conservative Republican Party isn’t a fluke: it’s at least in part a backlash against changing gender norms. And it’s likely to continue.
The analysis of the panel study looked at changes in attitudes on three different political issues: government aid to African-Americans, access to abortion, and support for marriage equality, controlling for factors like changes in employment, marital status, church attendance, as well as the age, race, and other demographic characteristics of the individual. Holding all these factors constant, conservative men who lost income relative to their wives became more conservative and liberal men who lost income relative to their wives became more liberal. For instance, views of abortion are measured with a series of seven questions, presenting different situations and asking if a woman should or should not be allowed to have an abortion under those circumstances. The mean Republican man who lost income (a drop of about 39 points in relative income) dropped by 0.3 points on the scale, while there was no significant change among those who gained income relative to their spouses. Similarly, Democratic men who lost income became more supportive of abortion rights: a loss of 39 points in relative income led them to go up on the abortion scale by about 0.4 points. Men who didn’t identify with either party didn’t move one way or the other in response to losses or gains in income. The results are even starker when it comes to the question of government aid to African Americans: men who lost income relative to their spouses became more extreme in their views, while everyone else became less so.
The findings aren’t all quite so clear-cut: loss of relative income didn’t have a significant impact on men’s support for marriage equality, perhaps due to the enormous overall shifts in attitudes toward marriage equality that occurred over the course of the panel. In addition, the absolute income of the household—how much money is coming into the household overall—doesn’t seem to have a consistent effect, even though some posit that men in lower-income households might face more threat from the loss of relative income. Despite these factors, the findings are consistent with the previous studies. Men who feel that their gender identity has been threatened, whether by specific threats like loss of relative income or more diffuse threats, like a society that they perceive to be biased toward women, are changing their political views in response. However, men aren’t all changing their views in the same direction, though it’s easy to understand why it would look that way. After all, in American politics, men are much more likely to be conservative than liberal, so if threats to male economic privilege make conservative men more conservative and liberal men more liberal, on average, it would make men, lumped together, slightly more conservative. It’s only when the analysis allows for conditional effects, as in the last two studies presented here, that the polarization effect becomes evident.
Why would a threat to traditional gender norms lead some men to be more conservative and others to be more liberal? The answer seems to lie in what researchers have called “hybrid” masculinities, in which members of privileged groups, like young, heterosexual White men, adopt attitudes and behaviors that seem to be at odds with hegemonic masculinity. These attitudes and behaviors can serve as an expression of identity for White heterosexual men whose identity is seen as the societal default or as a way for men to distance themselves from what they recognize to be male privilege. However, some, like sociologist Michael Messner, have argued that even these seemingly positive, progressive masculinity expressions have a dark side, as they tend to stigmatize masculinities displayed by less privileged groups and obscure continuing inequities. After all, if White men take leadership roles in movements designed to empower women or racial minorities, are we really fixing power inequalities in society?
Setting these theoretical concerns aside, these studies point toward concrete effects of increasing levels of threat to gender norms in the U.S. The most important legacy of the Great Recession of 2008-2009 may well be the way in which it tolled a death knell for men’s unquestioned economic dominance in American society. According to Bureau of Labor Statistics data, between 2007 and 2009, the percent of households in which women earned more than their husbands spiked three points to 29% and hasn’t receded much since; these are millions of households in which men were suddenly faced with direct threat to their gendered privilege. The studies presented here indicate that the men in these households became, on average, more conservative on all sorts of issues in response, and, in particular, less likely to support Hillary Clinton. In this view, an increasingly conservative Republican Party isn’t a fluke: it’s at least in part a backlash against changing gender norms. And it’s likely to continue: the transformation of the American economy away from jobs predominantly held by less educated men didn’t stop in 2010, but is expected to accelerate in upcoming years, as more and more truckers, cab drivers, and factory workers lose out to automation. We can predict that such changes will lead these men to become, on average, more conservative and more resistant to a female candidate like Clinton, especially one who challenges gender norms in the way she did. To the extent that there is a bright side, it appears in the seeming increase in hybrid masculinities in which men whose wives come to out-earn them adopt masculinities that don’t require the breadwinner role and move toward more liberal policies and candidates. Hybrid masculinities aren’t without their problems, but for those troubled by the current states of American politics, they may be our best hope for change.