Despite a broad field of qualified women and minority candidates, two white men are now leading the Democratic presidential field. Even after supporting women for Congress, why are Democrats shying away this time? Neil Visalvanich finds that neither party discriminates against women or minority candidates in congressional races, with Democratic Party donors actually favoring white women. But that may not apply to this year’s presidential race. Seth Masket finds that, when told that Hillary Clinton lost due to a focus on identity politics, white women are more likely to support men running on an economics message than women running on discrimination.

Studies:The Party’s Primary Preferences” and “You Had Better Mention All of Them
Interviews: Neil Visalvanich, Durham University; Seth Masket, University of Denver

Transcript

Grossmann: This week on The Science of Politics podcast, do parties favor diverse candidates or white males? For the Niskanen Center, I’m Matt Grossmann. Despite a broad field of qualified women and minority candidates, two white men are now leading the Democratic presidential field. Even after supporting women for Congress, Democrats may be shying away this time. Do both or either parties want diverse nominees? Are memories of Hillary Clinton’s loss deterring Democrats’ support for women in 2020? Today, I talk to Neil Visalvanich of Durham University about his just accepted American Journal of Political Science article with Hans Hassell, “The Party’s Primary Preferences.” They find that neither party discriminates against women or minority candidates in congressional races, with Democratic Party donors actually favoring white women. But why doesn’t that apply to this year’s presidential race?

I also talked to Seth Masket of the University of Denver about his working paper with Pavielle Haines, “You had Better Mention All of Them.” They find that when told Hillary Clinton lost due to a focus on identity politics, white women are more likely to support men running on an economics message than women running on discrimination. Visalvanich and Hassell find that neither party discriminates against women and minority candidates in their primaries.

Visalvanich: We examined the relationship between the two major national parties and whether they support, are more likely to support, minority and women candidates in the primaries for U.S. Congress. What we find is that the parties are actually mostly agnostic when it comes to promoting racial minorities. Democrats are significantly more likely to support white women, but that support does not extend to women of color. Republicans aren’t any more significantly or likely to support women than men, male candidates. A lot of the paper is that we don’t find that parties discriminate against these underrepresented groups.

Grossmann: The conventional wisdom that Democrats are the party of diversity doesn’t really hold up.

Visalvanich: The popular current conventional wisdom is the Democratic Party is really the party of minorities and women, especially in the last election, and the Republican Party is generally the party of older white men. This has actually been a bigger issue recently because the gender gap in voting and also in representation has really been dramatic in the last few cycles. I think I saw recently that Republicans … Democrats added a whole … In 2018, Democrats added a whole slew of women representatives into their ranks in that election. The Republicans, I think they either lost or they didn’t gain any women representatives at all. I think it was partly a function of them just sort of losing in general. But there is this sort of perception that if you’re a woman or you’re a minority, the Democratic Party is the way to go.

But we investigated specifically whether the parties are, in their actions, actually favoring certain types of candidates over others. And what we sort of find, at least with our current data, is that, really, parties are largely agnostic. This sort of challenges the conventional wisdom a bit. The Republican Party isn’t any more hostile to minority women candidates when compared to white candidates. The Democratic Party, they’re favorable to white women, but otherwise they’re pretty agnostic to minority candidates and women of color. The parties themselves sort of take a different … At least when it comes to the funding apparatus that the parties engage with, they take a different tack than the conventional wisdom.

Grossmann: But recent history might matter. Masket and Pavielle find that the framing of prior losses affects the next election. This time it might be ideas about Clinton and identity politics.

Masket: The real focus of the paper is looking at how post-election narratives work and change the political environment. That is, every election and certainly every presidential election is followed by some attempt to define what that election meant, what the lesson of that election was, particularly for the losing candidate or the losing party. They want to know what was the lesson of it? The real takeaway from this paper is just having that discussion, just making arguments about why Hillary Clinton lost in 2016 can change the political environment for the next election. It affects how people think about the next election, about who could potentially win and who couldn’t, about what the right candidates are to nominate.

In this case for our particular analysis, we found that this identity politics argument that was floated around a lot in the wake of 2016, this was the argument that Hillary Clinton lost because her campaign, and Democrats in general, focused too much on reaching out to women, to African Americans, to Latinos, the LGBT community and so forth, and not enough to just sort of communicating a general message for all Americans, and as a result of that, white men, white working class men generally, felt alienated. It didn’t reach out to them, and as a result, they turned to the Republican Party that was speaking to them.

Regardless of any truth in this argument, we found that just making that argument had an effect on voters. Among white Democrats, it affects who they like the next time around. In particular, it seemed to affect women, white, female Democrats. Generally, they were leaning toward nominating a woman next time around, nominating a relatively progressive candidate who cares about addressing racism and inequality. Hearing this argument made them more likely to abandon those goals, a little more willing to just say, “Okay, I’ll nominate a man. Okay, I’ll nominate someone who just has sort of a general economic message in mind.”

Grossmann: Some history suggests that prior narratives like these mattered for the presidential nominees.

Masket: The battle over the narrative is an important part of what a party argues about, that parties will use … Different factions within a party will make claims about why they lost so they can try and win future contests. For example, in the 1980s there were lots of debates among Democrats about why they lost three presidential elections in a row. For some, it was evidence that, “Oh, we’ve gone too far in the racially inclusive direction and we need to move more towards the white Southern wing of our party.” In some ways that’s how you end up with Bill Clinton as the nominee in 1992. There’s not a ton of literature in that area and we were just trying to develop that a bit more.

Grossmann: For Masket, this research came out of a larger project on how activists are seeing the election.

Masket: I’m in the middle of a book project right now that is more generally the Democratic Party 2016 to 2020. I’m interested in capturing a party in the act of making a decision, that is, figuring out why it lost the last election, coming to a decision about who it wants for the next time around. So this is obviously a big component of this. Part of what I am doing is talking to activists in some early primary states, in Iowa and New Hampshire and South Carolina and Nevada, and getting a sense of … talking to them from early 2017 and getting a sense of why they thought the Democrats lost in the 2016 presidential race. And then repeatedly talking to them over time and getting a sense of how they’re evaluating different candidates and who they like and who they don’t like. That part is still ongoing and I’m looking at a couple of different ways of examining factionalism within the party. But also part of it, I was just curious about looking at election narratives themselves and how they’re built and what kind of effect they have on activists and voters.

Grossmann: Visalvanich started his research by thinking about the representation gap, connecting his interests with Hassell’s data on party network support.

Visalvanich: My research, coming out of my PhD program, was looking at minority candidates and that’s sort of what my dissertation was about. That’s sort of what my research agenda was largely about and still continues to be. I’m still working on that kind of research. But as I sort of study candidates more, I actually grew more interested in representation. I’ve become more interested in this because I think that it’s clear that representation, specifically descriptive representation … What I mean by descriptive representation is sort of like having representatives who look like you or have qualities that are like you. So if you’re having a representative, if you’re a black voter, a woman representative of the, a woman representative if you’re a woman voter, that sort of thing.

These are extremely important for a lot of minority groups. One thing about America is that it has a very large representation gap. What I mean by representation gap is sort of a gap between … Racial minorities and women are significantly underrepresented in Congress and in our political institutions compared to their proportions in the population, the proportions in the electorate. I think that closing this gap and investigating why this gap exists and closing and is a big part of incorporating different groups into the political process, providing quality representation. It’s what a lot of groups care about. But usually, prior research is really focused on the candidate-level dynamics about this. Are voters discriminating against minority candidates? Are the minority candidates … is there a sort of a pipeline issue, which is are there enough candidates even out there and running?

That’s been sort of the focus on this representation gap. I was aware of Hans’ his work on on the roles parties play in sort of providing resources to certain candidates. We’re friends and we went to graduate school together, actually. We just started talking and we were like, “Yeah, this is interesting. What is the role that parties play in bridging this gap?” Then, so that got us thinking that we could combine my interest in representing minority representation, and Hans did all this work on party campaign finance networks and party influence in primaries and we sort of merged those two things together for this paper.

Grossmann: Visalvanich and Hassell used shared donors with the party Congressional committees to measure party support, and find that it matters a lot to who wins primaries.

Visalvanich: When we talked about party support, what we really mean is that the party as an entity has certain preferences and these preferences are expressed with support through access to resources. Obviously, in American politics, money is a huge factor in waging a successful campaign. So that’s a big part of it. But this also means access to adequate candidate training, staffing opportunities and other kinds of resources that make for a successful candidacy. Primarily, the way we measure party support is through shared donors between the candidate and the national campaign arms of both parties. What that means is that the party is sort of coordinating resources to be sent to certain campaigns and certain candidates. This is sort of seen as sort of like a … that they’re favorable to certain candidates over others. But really, the measure is meant to be kind of like a proxy variable for overall party support, that the parties sort of express their preferences this way.

In Hans’ work, he finds … He wrote an entire book on this and he wrote a number of different papers. He finds that this support is a significant part of the winning formula for successful candidates. It not only provides resources but it also has the effect of kind of clearing the field for favored candidates and signaling to other competitors who the party supports, and in and in some instances signaling to like the voters themselves that this is the candidate the party favors. The parties themselves kind of play a major role, actually play a major role in promoting diversity. What they do at this sort of candidate selection stage or this sort of primary stage, it matters a lot for the diversity of the candidates who make it to the general election and who have a chance to be elected.

Grossmann: The parties are interested in diversity but may perceive a trade-off with electability.

Visalvanich: There is a value in diversity for the national parties. I mean you can sort of see this in the way that the national parties really structure their conventions, right? I mean, I think that both parties really wanted to emphasize that, whether true or not, that they were diverse or they’re trying to be more diverse. I think that, at least at that level, the parties want to bolster their party brands by being able to claim that they broadly represent America. In a way, we’re kind of examining whether that rhetoric that both parties try to speak to, whether their actions actually match the rhetoric, in a lot of ways, I think. On the one hand, parties do have an incentive to discriminate against women or minorities because they may feel that they’re not as strong of candidates or there may be sort of a legacy of discrimination there that we’re not picking up.

But on the other hand, there is this sort of tendency … “Well, we want to bolster our ranks. We want to bolster our ranks with a more diverse array of candidates in order to make ourselves look good, to bolster our brand.” Those two things very well may be canceling each other out and that could be why we’ve observed the results we do. I want to talk a little bit about the speaking to results and the interpretation of the results, because I really think that you could interpret the results of the findings in two ways. On the one hand, we do find that the parties don’t discriminate. I think that there was a time in the not too distant past where I wouldn’t be surprised if the parties would have gone out of their way to exclude minority candidates.

I’m from Chicago. A lot of the ways, the Chicago machine was built to exclude African Americans for representation. I don’t think that that’s controversial or a secret. So on the one hand, you could sort of interpret these results as progress and a positive result, but on the other hand these results also show that the parties themselves aren’t going to go out of their way to help these groups either. Despite sort of big talk on both sides about increasing diversity, we’re not really seeing parties go out of their way to kind of give these types of candidates the resources that they need to wage competitive campaigns.

Grossmann: Minorities are still underrepresented as candidates, but there are women and minorities in each party.

Visalvanich: The most common candidate for political office in America is a white man. That is certainly what we find in our data. That isn’t surprising. But there were, at least in the elections that we looked at, which is 2010 to 2014, there were a significant number of minorities who ran under both parties. It wasn’t just all on the Democratic side, although the Democrats did have more black, Latino, Asian, and women candidates running, that a significant number in both parties are people of color as well. I would say that the main determinant for running, at least with regards to minorities is how many co-ethnics there are in the district when they’re seeking office.

In other research and what we looked at, I think that that’s sort of the determinant factor in terms of whether they actually seek office to begin with. I think, like white candidates, there are a mix of competitive and long-shot candidates among white and non-white male and women candidates. But I also think that … I think Democrats tend to have more competitive candidates, and this is probably due to a couple of things, that there are probably more Democratic office-holders who are minorities or women at the state level and they’re more likely to want to seek promotion.

Visalvanich: For minorities or women at the state level and they’re more likely to want to seek promotion as opposed to Republicans, so there’s sort of this pipeline thing coming up so you probably have less competitive candidates in the Republican Party.

Grossmann: Visalvanich says parties don’t force minorities into minority districts. Racial minority candidates instead self select.

Visalvanich: I think that there are a couple of things going on there. I think part of it is a supply issue. We mostly look at it at the congressional level, which is kind of like the top level of American politics, and often times, candidates move their way up through the state level, from the local level to the state level. Could be a lot of these minorities start from minority heavy areas, and that’s where they seek office. That’s where the pool of candidates exist for them to actually try to seek promotion. I think that that’s one part of it.

I think that part of it is also kind of a psychological part as well, in the sense that maybe candidates, they don’t feel like they could win in more white districts, or they don’t feel like they could get the support. Our results show that parties really don’t seem to care about those things, but even if there’s that perception out there, that could deter a lot of candidates from seeking office.

Grossmann: They do find that Democrats prize representation by white women. White women get twice as many party donors as white men.

Visalvanich: The results are really very dramatic for white women and Democrats. I say white women specifically because really our findings are the most significant or were only significant for white women. To put it in context a bit, the numbers of donors who donated to white women candidates in [inaudible 00:17:38] which is our measure of party support, is over double that of white men. It’s really a remarkable result.

I think that there are kind of two things that could be happening here, and I think that one of the things is that there could be a demand among primary voters for this kind of candidate. I think that what we see here is a result that is driven by an increased influence in these representation focused policy groups within the Democratic Party. Groups like Emily’s List, for instance. These groups have been a part of the Democratic Party apparatus or this sort of network as it were for a while, but really we’ve seen them become a lot stronger in more recent times. They’ve really exerted a lot more influence over all these things that we talked about, kind of like connecting candidates to donor networks.

Their specific demand is more descriptive representation of women in legislatures, in the congress, and they’ve been really effective at doing that. I think that our results really speak to how well that these sorts of groups have done for women candidates in the Democratic arty.

On the other hand, at least our results with the elections that we looked at, this support isn’t going to women of color, as well, which is sort of an interesting result. They really are more support for white women than women of color, and so that’s sort of a curious result that we’d like to dig into some more about why we’re seeing that.

Grossmann: Group organizing may be the key to women’s representation among Democrats.

Visalvanich: The influence of groups like Emily’s List shows how important the integration within the party of these sorts of groups is key to achieving descriptive representation goals. Given the importance of party support and being successful in the primary, especially at the more micro level, I think that one of the key implications of our finding, I think, is that in order for minority groups to really achieve what women have achieved in the Democratic Party, there needs to be more empowering of these representation focused groups within the parties, and not just the Democratic Party, but Republicans are consistently dealing with this issue as well, that the Republican Party has become largely a party of white men, and if they want to diversify, then the empowerment of these groups is key to that diversification.

Grossmann: They also find that party support helps candidates win the primary regardless of their demographics.

Visalvanich: We look at outcomes and party support, and basically what we find is that party support doesn’t matter more for minority women candidates when it comes to actually winning than other candidates. What that means is basically that party support works the same across a different array of candidates, so that if you get party support as a black candidate, it has the same effect as getting party support as a white candidate or latino candidate, for instance.

What that really means is that party support isn’t any more beneficial towards certain groups than others, it provides the same benefits across a different array of candidates.

Grossmann: Masovonich says the findings might apply to the presidential level, but they’re a different dynamics.

Visalvanich: Obviously, we look at a different level of politics than the presidential level, and I wouldn’t be surprised if the past, and Hillary Clinton faltering in 2016 wasn’t a major consideration among some Democratic Party elites. On the other hand, I think that there is a large segment of the party that is increasingly influential and that is willing to pay whatever electoral price they need to pay to get women candidates into office and women’s representation.

I’m not sure how this is going to play out at the presidential level to be quite honest. I think that there are sort of different dynamics going on here at the presidential level and at the congressional level. In a way, at the presidential level it’s a much bigger deal to voters and party elites, right, because it’s the biggest office. Those sorts of perceptions may matter more than at the congressional level where it’s more micro and I think that there may be less of an attitude that we need to get this one right as opposed to we really need to make a concerted effort to increase the degree of our diversity within congress and within our candidate pool.

The other difference that I could see there being is that I think that parties may have an easier time exerting their influence in certain ways at the congressional level. Obviously, there’s a big line of research about the party, the size at the presidential level, so I want to be clear that I’m not saying that parties don’t exert any influence at the presidential level, but I can see an interpretation of a lot of the research here is that party resources may be more influential at the congressional level, and they may be more better at clearing the field as it were. Those sorts of things may matter more at that level.

Grossmann: Masket and Pavia find that one frame around identity politics in 2016 might be effecting how the party decides in 2020.

Masket: There’s at least some evidence that this identity politics framework was a pretty common narrative. I don’t know enough to say yet whether it was dominant narrative, but if you look at media coverage and a lot of the punditry right after that election, this concept of identity politics comes up a lot, and it’s not just folks like Mark Lilla or Francis Fukuyama, it’s a lot of mainstream political reporters and pundits and a lot of different venues saying things like, “Well, it’s because she didn’t communicate properly to the white working class.”

You can see different versions of this argument. A lot of people have argued that, well, Hillary Clinton lost because she didn’t go to Wisconsin. In some ways, that is a claim about the campaign, that’s saying, “Okay, she didn’t run her campaign effectively, she should have known where the competitive states were and she should have made more of an effort to reach out there.”

That’s also in some ways an identity politics critique. That is, it’s saying, there were working class whites in that state, and that state flipped to Trump because the Democrats and Hillary Clinton specifically weren’t reaching out to their concerns and weren’t campaigning there.

Grossmann: There were a lot of ways for voters to encounter this narrative about why Clinton lost.

Masket: In many ways, voters learn about this the way they learn about all sorts of narratives in politics. They might read an article or read an oped. They might see some coverage on the news, or a friend of theirs will have posted something on social media that they’ll read. There was no shortage of punditry out there, particularly in the week or so after the 2016 election to be exposed to. I think the post-2016 election environment was particularly ripe for that sort of thing, just given that the presidential election had such a surprising and controversial result. Just given were polls were, a lot of people were very surprised to see Donald Trump win. A lot of Democratic voters were rather shocked, they were rather despondent, and probably paid more attention to the political news in the immediate wake of the election than they would after a more typical election, simply because they were looking for some sort of explanation, they were looking for some sort of story.

Grossmann: Their experiment found that the frame mattered most around women, perhaps because men had already heard it.

Masket: Figured there’d be some differences across genders, but we didn’t really expect to see pretty much the entire effect concentrated among women. Yet, in some ways that makes sense. This version of this identity politics narrative, that Hillary Clinton didn’t reach out to the concerns of white men, confirms what a lot of white men already felt, that the campaign’s going to have problems when it focuses on someone else. This confirmed a lot of what they had been hearing, so there wasn’t really a whole lot of movement there.

Why did it affect women? In many ways, we’ve investigated a fair amount of the literature on different socialization of men and women, there’s whole areas of other literature we could be getting into here and probably need to dig into even more, but there’s considerable evidence that women are generally socialized in the United States to be consensual, to seen consensus in their social groups, and men are socialized to be more individualistic. If women continue to receive this message that the party would have won if not for this candidate who was advocating for women, in some ways the first interpretation of that is, “Oh, it’s our fault. We wanted too much. We should try and scale back our needs on behalf of the group.”

That identity politics argument is in many ways really pernicious in this way. It taps into an area where women voters are already very sensitive. This concern that they’re in some ways asking too much by having a female candidate who’s focused on gender inequality.

Grossmann: It produced big moves among women. They became less likely to support women candidates, and less likely to support discrimination focused messages.

Masket: We used what’s called a conjoint study which is a way of dividing up if you have a lot of different variables. In this case we wanted to know how people felt about generic candidates who were either a man or a woman, who were either white or a person of color, or who were advocating for a general economic message versus an anti discrimination message and so forth. The conjoint study allows you to have several variables going at once, so just asking people a series of match up questions. Would you prefer a candidate who is a man, a person of color who advocates for a general economic message, versus a woman who’s a person of color who advocates for a generalist message.

You can change all those variables one by one so everyone in the study gets to see 10 different questions and we just see which they prefer. You compile those all together and you get a sense of actually how they rank these things. What we generally find here is that overall, and again, we were concentrating here on white Democrats, we found that white Democratic women generally preferred a female candidate, when we asked them who they liked, but we show them this identity politics narrative, and they move about five to six points more toward the male candidate.

They generally preferred a, given a choice between a candidate with a focus on the economy, rather than a focus on correcting discrimination in the workplace, but show them the identity politics narrative, and they move toward the economic message by about 10 points. These were fairly large effects, and those could be enough to effect which candidates they’re considering, or what policies they want those candidates to address once in office. Again, we found very small to negligible effects for men in this study.

Grossmann: Masket sees some signs this dynamic is mattering in the 2020 presidential race.

Masket: It’s extrapolating from one study we did of less than a thousand people. We can’t know for certain, but just looking at the field so far, it’s interesting. There’s a number of obviously very high quality candidates who are women and who are doing fairly well. Elizabeth Warren is in reasonable shape here. Kamela Harris is doing not poorly. Just digging down into some of the polls, a fair amount of the support for Joe Biden is his perceived electability, right?

The perception of electability is informed a lot by the lessons of 2016 that people are drawing. There’s certainly a fair number of people who are under the impression that Hillary Clinton lost because she was a woman, because of this message she had, and Joe Biden, by being a white man with some working class background and some ties to mid western region has a better ability to win. That’s not necessarily empirically demonstrated in any way, but that is a perception that people had drawn in some ways by these identity politics arguments.

Grossmann: It might even explain why there’s not yet much divide in men’s and women’s choices.

Masket: If this identity politics narrative that we were investigating has been taking hold, we would expect, as we found in the study, that the identity politics narrative in some ways encourages female Democrats to behave more like male Democrats. That is, they might have had more of an interest in backing a woman, they might have had more of an interest in backing a more liberal candidate with more of an agenda toward addressing discrimination, but because of this long standing concern that their identity concerns cost democrats the White House in 2016, in many ways their preferences are shifting, or at least their revealed preferences are shifting more toward what men were already thinking within the party, and so you don’t see as many differences across gender lines as you might have otherwise.

It’s possible had that view, that identity politics narrative never really taken hold, we might be seeing more gender splits within the party right now. We might be seeing more women coming out in support for Warren or Harris or Klobuchar or someone like that.

Grossmann: Masovonich agrees that presidential voters may feel differently than parties in Congress.

Visalvanich: I think that there may be kind of a psychological effect of losing, which could be a big motivator, and this might be happening at the voter level. I know I recently read a New York Times article about why so many African-Americans who very happily voted for Obama and were big supporters of Obama in 08, why are they supporting Joe Biden? And a lot of them are sort of like, “Well, we lost, and we just want to win, and we just think that a white man is the kind of person who people want to elect.” And I think that at the very least that these dynamics are super interesting and deserve a lot more exploration, the sort of psychological effect of losing.

Visalvanich: It is at odds with kind of what our paper finds, which is that a lot of the donor class at the very least seem to be pretty motivated for more representation kind of spurred on by these very influential interest groups in the party, and it’d be interesting to see whether this sort of holds into 2020 where Trump is going to be on the ballot, and different dynamics may take hold, and that’s definitely going to be interesting to look at.

Grossmann: Traditionally, Masket says, party chairs feared women and minority candidates, but now that might be changing.

Masket: We may be talking about two very different things here. There is some research by I think Michael Miller and others, surveys of local party chairs, Democratic party chairs, finding that they’re generally uncomfortable nominating a woman or nominating a person of color thinking that white men are more electable, and there’s I think some other research by Kira Sanbonmatsu found something similar to that a few years back.

It may be that that’s shifting, particularly in the wake of 2016 as you have sort of a new activism by a lot of women in the party, a lot more women are running for office now, and it may certainly be that within the 2018 context, female candidates were seen as having an … And this was seen as a place that was very important to a lot of Democratic activists, and so that may be a relatively new thing within the party. I don’t know really that the party is unified in this.

I mean, I’m speaking somewhat in generalities here, but it may simply be that as some parts of the party, some activists, some voters are feeling very passionate about nominating women, nominating people of color for high office, there may be others in the party, some voters, some party leaders, some people with much more history in the party, who are still wary of it and who drew a lesson, an identity politics lesson from 2016 that was similar to lessons the party drew in previous contests

Grossmann: 2018 saw a big upsurge in representation in Congress different from the presidential race.

Masket: An activist base of women that has cropped up since Trump’s election and particularly since his 2017 inauguration when you had just record turnout in political protests in January 2017 on specifically a message about women. There are just a lot more female candidates. There’s a lot more women activists now on the Democratic side, and just looking at representation-wise both in Congress and in state legislatures, you have record numbers of women in Congress and in many state legislatures right now but almost entirely on the Democratic side.

In many places, the number of female Republican elected officials has actually shrunk somewhat in recent years, so women’s political activism has become increasingly a party story, and I feel like Democratic elites not only are many of them women now, but many of them are sort of recognizing this as a very important and very vibrant vein to be tapping into and realizing that there are political risks to ignoring this aspect of representation and activism.

Grossmann: Visalvanich also sees gender dynamics changing. The Republican party may have long odds to generating women and minority candidates in the Trump era.

Visalvanich: Even if the party is trying, which they are, which they are really proactively trying to promote nor are they denying resources to minority women candidates, but I think that the Republican party has a major Rand problem with regards to how people see the party as whether they’re receptive to candidates who are minority and women, and I think part of it is sort of policy, part of it is sort of kind of the president himself, and so kind of president Trump, and his relationship with these different groups has not been good to say the least. And it’d be interesting to see kind of in the era of Trump how this all plays out, but maybe this sort of shows that even with the party being kind of at least agnostic in the way that it spreads its support around different candidates, that if you can’t get people to run as Republican, it doesn’t really matter, does it?

If you can’t get minorities and women to run as Republicans, it’s pretty hard, and I wouldn’t be surprised if in the Trump era, we sort of saw the Republican party become at least seemingly to be a lot more hostile place for those kind of candidates than before.

Grossmann: He says even though they didn’t find discrimination by either party, there still might be minority disadvantages at other stages of the process.

Visalvanich: We examined just one part of the electoral process, which is the primary, and that its entirely possible that discrimination can happen in other stages of the process, like the recruitment stage, for instance. I mentioned this with regards to Republicans. If people don’t want to run under your banner, and that could be for a number for reasons, that they either feel like your party brand is hostile to their kind of subgroup, or they feel like the party apparatus itself isn’t very favorable to them, rightly or wrongly, they’re not going to run.

And so the kind of discrimination that we don’t observe here, we might observe in other contexts is what I would say, and the other point I’d like to make is that this paper kind of exclusively looks at candidates who have made it to a certain point in the process, so it could be that the quality of minority women candidate who do make it to this stage, who make it to a stage where they’re running and they’re raising money, is different from the quality of white candidates and that this quality isn’t something that’s picked up in our control variable on candidate quality, which is whether they held prior office.

Grossmann: Visalvanich says lots of things still need to be done to increase representation by proactive parties.

Visalvanich: Our results don’t explain the representation gap, but they do potentially point to a way to kind of help close the gap, and to be clear, I’m not saying that parties need to do everything. There’s obviously a lot of factors involved in under-representation in America, including pipeline issues, whether these under-represented groups feel like the parties are welcoming, whether they think they can win, whether they actually have a means to move up from the local level to the national level, and there’s only so much parties can do.

But I truly believe that parties can be more proactive and in a way have to be more proactive in diversifying their ranks, and whether this be through the empowerment of different groups or whether this be kind of from a top level decision at the committee level, I think that it’s a key part of the equation, and I think that one of the things that these results sort of show is that parties, they themselves probably are agnostic, and they’re not motivated, so they kind of need to be pushed to be motivated about these sorts of things.

And I think that you can sort of see with these results that these results happened. We looked at these results in 2010 to 2014, and in a way, you can sort of see that this culture of promoting women in the Democratic party has really borne out in the elections that we’ve sort of seen recently, that we saw way more women be brought into Congress, and I think that these results reflect that desire among Democratic party donors and elites that they want more women in Congress, and I think that if we want to see more diversity in our political institutions, parties are going to be a big part of that equation.

Grossmann: Masket says the party’s interpretation of elections currently hurting women candidates might not last as party elites see how voters are changing.

Masket: There was a Democratic campaign activist I spoke to in New Hampshire, this woman who’s worked on a lot of presidential campaigns in the past. She worked for Hillary Clinton both in the 07-08 cycle and in the 15-16 cycle, long-standing, lifelong feminist. I had asked her in 2017 why Hillary Clinton lost, and she said she was concerned it was gender. She said, “I’m concerned that we need to nominate a white man next time around. That’s the only sort of candidate who can stand up to Trump,” and she said, “I’m worried I’m going to get kicked out of the women’s club because of that.”

And I followed up with her a few months ago about that, saying, “Right after the election, you said the following,” and she said, “Oh, I remember what I said. I don’t agree with that anymore. Just watching what happened in the 2018 cycle where a lot of very strong women, a lot of people of color won, I don’t think that’s actually true.” She isn’t committed to a candidate yet, but she has not eliminated women of people of color for that role. It is interesting to me how these narratives, they do affect the way people think about politics, but they’re also not necessarily permanent, so people can re-evaluate as things go forward.

Grossmann: There are lots of views of 2016 after all, and it’s hard to determine which, if any, make sense.

Masket: This election had so many different possible frames. This is one thing I talk about when I’m interviewing political activists for the study I’m doing. First question I always ask is, “Why do you think Hillary Clinton lost in 2016?” And that’s often an hour-long response to that. I mean, people have a lot of views on that. That’s a whole wonderful conversation, and I’ve gotten a really wide range of interesting answers that have come out. Some people talk about identity politics. Some people talk about campaign deficiencies or candidate differences.

Some people will talk about Jim Comey or Russian influence. Some people will say it was about Sanders supporters, or it was about Jill Stein. It was a close enough election that almost any of these answers are potentially right or potentially wrong. It’s also close enough that you just sort of take the political science approach that it’s very rare for a party to hold onto the White House for three consecutive terms. You had middling growth in the economy. That’s a situation that’s probably going to be a 50-50 election, and so maybe we shouldn’t worry so much about what the candidates and campaigns and parties were doing and just say, “Well, this was always going to be a toss up, and it ended up something like that.”

Grossmann: Next up for Visalvanich is to see if anything is changing under Trump.

Visalvanich: Parties are constantly changing, and as I mentioned before, there has been this dialogue in the GOP very recently, especially after the 2018 election, that, “Well, we have this major problem, especially with the gender gap, and we need to diversify our slate of candidates,” and they seem to have this sort of conversation with themselves over and over every time they have a bad election basically, so in 2012, they kind of had the same thing.

And so it would be interesting to see whether how parties really have changed in the era of Trump, and so once we have the data together, we’re going to want to see whether the relationship between, for instance, the Democratic party donor networks, whether they’ve become more receptive to women of color or whether they’ve sort of stayed the same, whether Republicans actually are trying to reach out to minorities or whether they’ve totally given up in the era of Trump. And so I think that one key thing to note is that because parties are constantly changing, we shouldn’t expect these results to sort of stay the same. We actually should expect the results to change.

Grossmann: Masket will also be following up on the primary and then the general election.

Masket: One of the things I’m hoping to do is we might do some follow up surveys on this and possibly do a follow up study right after the next election. One of the problems with the study as we have it is that it occurred well after the 2016 election, and we might want to be able to do some sort of a study before post-election narratives take hold, so that would be kind of interesting.

I’d like to see if we could ask more explicitly questions about electability and what makes people decide one candidate is more electable than the other, and generally, for my book project, I’m going to be continuing this interviews and then sort of seeing how people respond to the debates and how people respond as the campaign heats up, things get a little more negative, then we start to move into the voting early next year.

Grossmann: There’s a lot more to learn. The Science of Politics is available bi-weekly from The Niskanen Center. I’m your host, Matt Grossmann. Thanks to Neil Visalvanich and Seth Masket for joining me. Please check out The Party’s Primary Preferences and you had better mention all of them, and then listen in next time.