Joe Biden is about to select his vice-presidential running mate, having pledged to choose a woman. Will the pick change his chance of victory or the future of the Democratic Party? New research suggests running mates may not have the direct influence that most expect—but they do send strong signals about presidential candidates and their parties. Christopher Devine and Kyle Kopko find that vice presidential nominees do not have home state or regional effects and do not seem to help attract affiliated social groups like women. But popular running mates can rub off on nominees’ popularity and change how the voters see them ideologically. William Adler and Julia Azari find that running mates are a party decision; parties try to balance their ideological coalitions, with a variety of inputs from public and elite co-partisans. Biden’s pick will help define the candidate and his party.

Guests: Christopher Devine, University of Dayton; William Adler, Northeastern Illinois University
Studies:Do Running Mates Matter?” and “The Party Decides (Who the Vice President Will Be)


Matt Grossmann: This week on The Science of Politics. How vice-presidential running mates, change parties and elections. For the Niskanen Center. I’m Matt Grossmann. It’s beef steak season. Joe Biden is about to select his vice-presidential running mate. Having pledged to choose a woman, what criteria will guide his decision and how will it change his chance of victory and the future of the Democratic Party.

New research suggests running mates may not have the influence that most expect, but that doesn’t mean the choice is irrelevant. It may send strong signals about Biden and the State of the Democrats. Today, I talked to Christopher Devine of the University of Dayton about his new book with Kyle Kopko, Do Running Mates Matter?

They find that vice-presidential nominees do not have home state or regional effects, and don’t seem to help attract affiliated social groups like women. But popular running mates can rub off on nominees’ popularity and change how the voters see them ideologically as well as their readiness for office. I also talked to William Adler of Northeastern Illinois University about his paper with Julia Azari, “The Party Decides (Who the Vice President Will Be).”

They find that running mates are still often selected by a party decision. Parties try to balance their ideological coalitions with a variety of inputs from the public and elite, but each party in every case is a bit different, with seriousness and balancing usually prevailing. Devine and Kopko look at many potential electoral effects of running mates, finding some real influence, but also some popular myths.

Christopher Devine: We’re trying to really establish whether running mates, vice-presidential candidates matter. And of course, it’s a topic that we’ve heard about throughout American history. We hear about it every election cycle, but there’s never really been a whole book devoted to this question of do they actually matter in the sense of influencing the outcome of an election? You know, do they actually move votes essentially? And that’s what we’re trying to tackle here and bring together a lot of trans of that.

So in particular, what we do is we focus on three types of effects, three different ways that running mates could influence the outcome of an election. So we have chapters on each of these. So the first one is direct effects. That is, do running mates, their popularity, opinions of them, maybe their unpopularity. Does that influence voters in general, just across the board, are people reacting to the running mates, popularity or unpopularity?

The second type of effect. I mean, these aren’t mutually exclusive, but they are different ways of thinking about it. The second type is targeted effects. This may be the most familiar one. This is the idea that, it’s not that you’re going to really persuade voters in general with a VP pick. Well, you might be able to pick up support among a certain group of voters. So this year we have Joe Biden has had said, he’s going to pick a woman as his running mate.

Of course, we had a couple, just two women running mates in the past. So we could look at what effect does that have on voting among women or we analyze this among for instance, Evangelicals and Catholics and working with what’s available. Of course, we haven’t had an African American or Latino running mate for instance and we may see that this year.

Working with the evidence that we have, what effect does that have? The third one? And this is where we find the most significant effect, because I should say, I guess I didn’t really. We don’t find much in the way of direct effects. There are some people seem to be swayed by the running mates popularity. I can get more into that. I’m sure in this conversation, but it’s pretty minimal.

And to the extent that there’s any spike in the running mates, popularity that effect fades out pretty quickly and probably doesn’t end up influencing voting a lot. Targeted effects as far as picking up groups of voters, we find very little in terms of winning over women for our own Palin, or winning over religious groups, really the only target effect we find is that Paul Ryan seems to have picked up votes among conservatives in 2012.

We find some evidence of that, but it’s all pretty minimal. So where we find the most effect, I think this is intuitive. It’s not the first time anyone’s expressed this idea, but there’s never been robust evidence of this. There are some studies in the liternature, really, two, that get at this effect, these indirect effects, but nothing really comprehensive.

And that’s what we do here, is the last one we establish is indirect effects. And that is that basically people are still voting for the presidential candidate. They’re not voting differently because of the running mate. There’s some direct reaction like wanting to elect their favorite vice president over someone they prefer less for president. The effect is that running mates shape perceptions of the presidential candidates.

So the fact that, say Joe Biden picks someone… If he picked someone who was clearly unqualified for office, and the analogy here would be to John McCain in 2008, maybe that’s too harsh on Sarah Palin, but at least there were significant questions for her qualifications.

If Joe Biden were to pick someone who’s clearly unqualified, our evidence would suggest that people would have so a less favorable view of Biden’s leadership skills, his judgment let’s say, and they might be less likely to vote for him as a result. Not because they’re trying to keep someone away from the white house as vice president necessarily, but because they’re doubting whether Joe Biden is up to the job.

Contrast that with if you were to pick someone who’s well qualified, that reassures people, this is someone who’s going to make responsible decisions and really try to act out of the nation’s interests rather than his own interests. We also find other effects for ideology, which I’m sure we’ll discuss that the running mate can shift some perceptions of the presidential candidates ideology.

So all in all, we’re not talking about a dramatic effect on the race, but there can be an important effect on how people are evaluating the presidential candidates and choosing between them because of how they make this, what is often called First Presidential Act.

Matt Grossmann: Pundits like to talk about running mate influence, but most still doubt that they matter.

Christopher Devine: They wanted to stage something of an intervention with folks with pundits, with academics, a lot of folks who, as we document in the book, I’d say they’re conflicted about the role of not just vice presidents, which we also talk about, but also running mates. They’ll often dismiss it as a parlor game, right?

You’ve heard that phrase, this is just something we do every four years to entertain ourselves and often kind of sneer, well, I know and everybody knows that it doesn’t really matter. And yet you track the same folks a little while later, again as we document in the book with a few examples, and some of these folks will then talk about how picking a certain running mate could change this race.

You know, oftentimes I think there’s, and this would be to your point, that’s sort of the conventional wisdom. It’s hard to say what is, because on the one hand, a lot of people are acting as if… Well, we all know running mates don’t matter that much, but then when we start strategizing as political junkies about how to win the race, there’s tendency to really overState what effect the running mates may have. And that’s one of the things we’re trying to document in this book, but also to test it out, to see can we really get comprehensive evidence on whether this matters

Matt Grossmann: Adler and Azari analyzed the selection as a party led process?

William Adler: The paper that I did together with Julia Azari of Marquette University, is sort of a deep dive into vice-presidential selection and party politics. And what we did here is we looked at, instead of focusing only on individual nuances or quirks of the vice-presidential selection process, we wanted to take sort of a bigger picture perspective on what does it look like from the perspective of the party when the presidential nominee chooses their running mate.

And so what we do is a couple of different things. One is, we look at first of all, sort of building on studies that have been done previously of vice-presidential selection, reframe how we think about some of the variables that go into that especially ideology. Because of course there is this sort of longstanding understanding that run people who are running for president, they want somebody on the ticket who maybe balances them ideologically.

Is one way to think about it. And one of the things that this literature up until now has, has looked at is saying, okay. Well, these people are, they’re more liberal, they’re more conservative, they’re more moderate, but the way the studies were done was kind of too vague of what they mean by ideology.

And so we use the DW-NOMINATE scores or other DIME scores where possible to try to get out of a better look at what ideology is as a continuous variable to see if that’s true, if there is this ideological balancing going on or not. And then what we do is do a deep dive into a couple of case studies of vice-presidential selection based on archival research. One about the selection of Bob Dole as Gerald Ford’s running mate in 1976, and then the other of Dole as the presidential in 1996, choosing Jack Kemp to be his running mate.

Matt Grossmann: They confirm that parties seek balance, but that shorthand may mess partisan differences.

William Adler: I think the conventional wisdom, like I mentioned previously, is the residential nominee wants someone who balances the ticket in some meaningful way. And obviously in the past that often meant a geographic balance. If you think about the 19th century parties, a Northern presidential candidate and a Southern running mate, or vice versa, or an Eastern candidate and a Western candidate or things of that nature. And more recently you think about it in terms of ideological balancing.

Well, if the presidential nominee is more liberal, maybe you choose a running mate who is a little more conservative or vice versa to pacify some part of the party that hasn’t been represented. And we think that has a lot of validity to it. But then, again, there’s a lot of nuance that’s being missed by that. One thing that we find, which I think is important, is that there’s not necessarily the same sort of behavior in the Democratic Party and the Republican Party.

We find an asymmetry there between the way Democrats choose their running mates and the way Republicans do, which is to say, we find Republicans tend to choose running mates significantly more conservative than the presidential candidate. Whereas for Democrats, they don’t necessarily choose running mates who are more liberal than the presidential candidate. In fact, in many cases, they choose ones who are significantly more conservative.

And so you have this pattern of Republicans playing to the conservative base of their party, but Democrats don’t necessarily do the same thing. And so that’s one important thing to look at. The other thing which we think is important that doesn’t always get captured in the conventional wisdom is that choosing a running mate is not only the individual choice of the person who is the presidential candidate, but it really has to be thought of from this broader perspective as the party helping to make that choice.

William Adler: Ultimately, of course, that one person does make the final decision, but the factors that go into that and the interplay of the different forces within the party seem to have a lot of bearing on how those choices are structured.

Matt Grossmann: Adler says, running mates may be becoming more important over time.

William Adler: One of the things we find is that the importance of the vice presidency has grown a lot in contemporary times, obviously the recognition of how fragile the presidency could be, passage of the 25th Amendment and then the party reforms of the seventies. I think really sort of highlight the need to have somebody there with you who can really be a governing partner. And that’s where Jimmy Carter’s selection of Walter Mondale comes into play in 1976.

Carter was sort of the first person to run what we would now consider the standard vetting for a running mate, but hadn’t really been done in a thorough way prior to that time of doing these interviews, background checks, sort of really deep dive into this person’s life before you make them your running mate.

Some of that of course is also a reaction to the selection of Eagleton by George McGovern in 1972, and sort of failure of that nomination ultimately and having to be replaced, because of some of the issues that came up there. And so Carter goes through this really thorough, very public interview process and lands on Walter Mondale as his running mate.

And Carter and Mondale come to this agreement, that Mondale will be a full governing partner, not just somebody shunted off to the side, once Carter gets elected. And that really changes the character of the process to think about that person as someone you will actually work with, not just for the sake of getting yourself elected, although of course that too. And so I think the level of seriousness is more important, but also what you see is that it sort of highlights just how central it is to have your entire party on board.

And one of the things that you find, as I mentioned before, is that because Republicans are picking these more conservative running mates, it’s not so much a question of is this a faction of the party that needs to be appeased, but where is the center of gravity in the party going to land? And that’s really seems to be the function of that running mate choice, is to help sort of not balance per se, but situate the presidential candidate within that party center.

Matt Grossmann: Devine says, the most commonly studied outcome is geographic electoral effects, but that’s where the studies find little influence.

Christopher Devine: There’s a fairly lengthy literature on the home state advantage, or you could say home region advantage, usually it’s home state, but sometimes region. When we tackle that in the book, we look at the home state effect and also the regional effects. Most of that literature has shown that they’ve not found any statistically significant home state advantage on average.

It’s not to say it never happens, but at least as a general matter, where previous studies have landed and where most of our studies have landed there’s not much of one. But, a lot of them starting with Dudley and Rapoport, in 1989, they had a great article. They found that it was more of a conditional effect. Yes, not every running mates can deliver a home state advantage, but maybe once it came from smaller States. People know each other better there.

They know their politicians better. There’s more of a coherent identity. We picked up on that in our first article, and added the idea of experience. It’s not just coming from a small State, but it’s actually being experienced politician in that State. That’s how you get that kind of affection or loyalty that would lead to home state advantage.

And this is kind of where we went with our book of 2016 book, The VP Advantage. And we analyze that from a lot of different angles, kind of going beyond what we did in the articles. I mentioned the Lewis-Beck and Rice equation, that’s an intuitive approach. It’s basically looking at how much did the State in that year deviate from its normal trends in comparison to how the nation deviated from its normal trends.

There’s a kind of response article to that. Heersink and Peterson and they challenged Lewis-Beck and Rice equation. Some issues dealing with that and we have our response in the 2019 article that kind of gets into the weeds on that. So I’d encourage people to check that out. We kind of try to sort through, what are the differences here?

One of the issues was that their article came out about the same time as our 2016 books, so there wasn’t much of a conversation between them. That’s what really trying to do with the 2019 article. The brief story there, this 2019, this is in presidential studies quarterly. I’ll try to do the brief version of it. Is really our argument is, you’re right, there are some problems with Lewis-Beck and Rice equation, and we did use other methods in our book, but again that came out at the same time.

Can we address that issue? And also our other concern, which we express in the 2016 book, which is, if we’re talking about how people vote, the relationship between where they live and how they vote, that’s an individual level relationship. And traditionally and including in some of our previous work, most previous literature, the Heersink and Peterson piece, they use aggregate data for that, which is great when that that’s all you have.

But if there’s a choice to use individual level data, survey data to examine this individual relationship, and that’s preferable to use that to avoid committing ecological fallacy. And so that’s what we try to do in the 2019 article. Of course, the challenge is when you’re trying to look… This case we do basically regression analysis, try to predict whether people are going to vote for the democratic or the Republican candidate with one of the independent variables being survey respondent.

Do you live in the running mates home state? We use that as an independent variable and see whether it has a significant effect on how they vote. We find that it doesn’t. Now I should say this is only the 2008, 2012 and 2016 elections because it’s hard to find surveys out there where you can get a lot of people from Virginia or a lot of people from Indiana, let’s say. But like many political scientists, we have the joy of you know, getting into the CCES, Cooperative Congressional Election Study, massive sample about 60,000 in 2016.

And so you can get 2000 respondents from Virginia or 1000 from Indiana and see with pretty robust sample. Is there in fact any difference that comes from being someone who lived in Indiana or Virginia in 2016? What we find in both those cases is, no there wasn’t. There’s no statistically significant effect.

And in fact, what you do in the CCES data, we capitalize on some ANES data, in those cases when you account for people’s home state, it does absolutely nothing to improve the predictive quality of the model. So this is three elections. I want to be duly cautious here. I want to respect other work that’s been done, which I think has really advanced the conversation. I appreciate that.

I just think where we are is pretty clear to us at least that when you’re analyzing this individual relationship it’s best to use survey, individual level data and at least when we are able to use that in the most recent elections, there’s just not much there in terms of home state advantage.

Matt Grossmann: Another common theme, is running mate diversity, but Adler says that’s relatively new.

William Adler: Until the choice of Geraldine Ferraro in 1984. As the Democratic Party’s running mate. That was the first time a woman had been selected. We’ve found evidence, it’s not the first time a woman was necessarily considered speaking about the 76 case. And one of the names that was tossed out to be Ford’s running mate was Anne Armstrong who was the US Ambassador to Great Britain.

And they pulled about her and they got a lot of receptive, positive feedback when they were talking to people about the potential of nominating her. And we found a memo from Ford himself saying basically this is all well and good. I don’t think the country is ready to have a woman be the vice president yet. And sort of vetoing the idea just based on that. Other than the choice of Ferraro and Palin, no other women yet.

This’ll be the third time a woman will have been nominated to be the running mate. And that’s interesting in and of itself. And sort of the question of diversity broadly speaking, there just honestly hasn’t been much diversity in the vice presidency up until now. We found very little evidence in the case of Johnson in 64 or whatever, thinking about it from that perspective, even though there were people who were urging him to do it.

Ford got a couple of letters from people saying, why don’t you choose Senator Ed Brook of Massachusetts who was African American. The fact that it’s become such an explicit criteria here and such an interesting focus of consideration for Biden, I think is representative of where the Democratic Party is today.

And the fact that you had in the beginning of this electoral cycle, such a broad range of diversity of people who were running in the first place. And Biden sort of acknowledging very publicly that that’s what the future of the party will look like. And he’s kind of not what the future of the party is. And that’s why he keeps saying things like he sees himself as a transitional figure, bringing these new people onto the stage.

Matt Grossmann: Devine and Kopko find little evidence that selecting a woman has influenced voters in the past.

Christopher Devine: So let me start with the women running mates. We have Geraldine Ferraro, 1984 and Sarah Palin, 2008. Of course, there’s a lot better survey evidence available in 2008 and that’s where a lot of our book focuses. Yeah. One thing we can do is to look out, we use ANES data. So from 52 to 2016 and we can plot in that case when we do in the book, what’s the effect of gender controlling for various other factor is kind of standard battery of controls.

What is the effect of gender on vote choice across the years? And you would expect that if gender is influencing vote, basically more women are voting for the ticket with the woman running mate on it, Ferraro or Palin. You see the spike in the effect of gender on a vote choice in 84 and 2008.

Well, you look at that plot in the book and there’s just no significant difference. It overlaps with every other surrounding fact. Of course, we know within margins of error. So there’s no clear relationship there, even better. In terms of quality of evidence, we were able to use some panel data from 2008. So we get the same respondents in the sample, the survey research coming back to them again and again, and we’re able to track the changes in their opinion over time so we can see okay.

What was the effect of gender unintended vote choice, say in June, 2008? Well, before Palin was selected on August 29th. And we can see then if we control for those preexisting opinions is capitalizing on Gabriel Lenz’s research from his book. If we can use that method and control for preexisting changes in opinion based on gender.

Do we see some subsequent spike in the effect of gender and later stage of the campaign after Palin was picked. And in fact did her selection kind of activate the effect of gender on intended vote choice? And we just see no change in terms of women becoming more likely to vote for the Republican ticket. Of course, those are two cases, we’d love to have more evidence and we’re obviously going to have some this year to work with. But based on what we have, we just see no evidence when running mates increasing votes among women.

Matt Grossmann: Adler and Azari looked closely at the selections of Bob Dole in 1976 and Jack Kemp in 1996 finding a lot of concern with ideological balancing.

William Adler: So 1976 is an interesting case because you have Gerald Ford who is the president, obviously hasn’t been elected on his own merits, comes into the office of the vice presidency and then takes over after Nixon’s resignation. He appoints Nelson Rockefeller to be vice president. And there’s a lot of dissension within the Republican Party over that choice, because Rockefeller is a relatively liberal governor of New York and obviously famously had been opposed to Barry Goldwater and the conservative turn in 1964 within the party.

And so almost immediately you have a lot of pushback from within the Republican Party to Rockefeller. And it’s to the potential of Rockefeller remaining on the ticket for the 1976 campaign. Ford is also challenged on the right by Ronald Reagan for the Republican nomination.

And one of the things that happens in the course of the campaign is Ford eventually decides after a lot of internal pushback and a lot of back and forth with his advisors in The White House. He makes this very public statement that Rockefeller will not be his running mate in 76. When it comes down to it, Ford has most of the delegates, almost the majority going into the 76 Republican convention.

Reagan makes a play, try to grab some delegates from the Pennsylvania delegation by naming Senator Richard Schweikert as his running mate who’s more liberal. That play doesn’t work for it and has the nomination. And then at the convention, Ford and Reagan basically have a conversation, and Reagan says, “Well, I don’t want to be your running mate.” And Ford says, “Well, who would be acceptable to you?” And Reagan basically says, “Of the names that I’ve heard mentioned, the one that I would be okay with is Bob Dole of Kansas.” And Dole ends up being the nominee.

One of the things we found through digging through the archives at the Ford library is that Dole in fact had even been on the list of people that the Ford campaign was initially considering. His name sort of appeared very late in the process. They asked these brand polls about who would you want to be the president’s running mate, Reagan, Rockefeller, John Connally, Chuck Percy of Illinois, Howard Baker, lots of other potential names and Dole’s name wasn’t there.

And it came sort of late in the process. And what the Dole choice shows, I think is that, there was all this pushback within the party to the more liberal Rockefeller and Ford kind of felt that with the Reagan challenge and how close Reagan had come to taking the nomination, that he kind of heavily had no option other than to pick somebody who was on the more conservative side of the party.

Dole in himself when he becomes a candidate in 1996 our second case study then, ends up choosing former congressmen and Secretary of Housing and Urban Development Jack Kemp, who is a well known tax cutter conservative, very much sort of an icon of the conservative movement in this period, and sort of representing that more conservative turn in the party to the point where by 96 Dole was perceived basically as a moderate in the Republican Party compared to the Gingrich revolution and everything that’s going on in that period.

And again, Kemp isn’t choice who comes out relatively late in the process. There’s this long short list which we came across and the archives at the Dole library of the people they were considering. And Kemp’s name didn’t make the list of that initial 25 people they were thinking about. And one of the things that appears here also in this process is the sense that Dole is a little too moderate for a lot of the party base. That they really want somebody who’s more part of the conservative movement in the way they perceive it. Part of the the Reagan movement in the party.

And so some of the names that the traditional names that they consider people like Don Nickles, who is a Senator from Oklahoma or Carroll Campbell, former governor of South Carolina, John Engler of Michigan. They’re all fine. And Dole says, well, that’s all fine, well and good, but maybe we need something a little more. In some ways, exciting.

And we came across quite a lot of evidence that the Dole campaign was paying attention to, what conservative thought leaders on the outside journalists and people in the media were talking about who Dole’s running mates should be. And several of them had the same answer, which is that it should be Jack Kemp, Mona Charen, William Buckley, Bill Bennett. All wrote columns very publicly urging Dole to choose Kemp as the running mate.

And in the sense of if you think of the party a little more loosely, including these activists, these media figures on the outside within the conservative movement, Dole seems to be very responsive to them, and responsive to what this conservative base wants and sees himself as maybe somebody who needs that little jolt of excitement from the right wing of his party, from the base of the party.

Matt Grossmann: And Devine and Kopko find that voters also say they want balancing, and there’s some evidence of ideological effects.

Christopher Devine: As far as the effects of ideology on the voters that kind of balancing, I should say one thing we find when you look at polling about how people think about running mates, which you have to be careful because people, what they say in a survey and what they do in terms of their behavior can be two different things. But people tend to express a preference for balancing of some kind. They don’t express a lot of interests and sort of the electoral strategy stuff.

Picking someone from their home state or something like that, but they do express some interest in picking someone who does kind of balance a ticket ideologically. Especially in terms of experience I should say. And of course, we see candidates throughout the years often, not always look at Bill Clinton, Al Gore but often looking for someone who balances the ticket.

I think Michael Dukakis and Lloyd Bentsen for instance, or even Romney and Ryan or McCain and Palin. Going back to what is the big finding of our book? That’s really the indirect effects that matter most. Okay. I would question whether you’re going, let’s say by, in this case you have some people who don’t like him now, they’re still not going to like them later, but they’re thrilled that he announced Elizabeth Warren as his running mate.

And so now they’re going to vote for him even though they have some other kind of preferences in terms of presidential candidates, where maybe they would’ve gone or a party or not vote or something like that. I don’t think there’s a lot of that. What it does by picking someone who’s ideologically different is, I think it changes how you think about the presidential candidates’ ideology.

So let’s say John McCain’s, is of course the best example here, he originally wanted to select Joe Lieberman as his running mate, right? This is one of his best friends and politics in the Senate. And they traveled together and Lindsey Graham and three Amigos. On the issues that were most important to McCain foreign policy, the Iraq war, Afghanistan and so on. Him and Lieberman, they were just really well aligned. When it came to economic and social policy.

Joe Lieberman was essentially a standard liberal Democrat and if he had picked Lieberman as he wanted to, of course the party pushed back, and this really, I’m sure it gets to Julian Williams for research, if the party had, and the party did pushback some kind of early signals of that. And that’s why McCain and his team had to make this rushed pick at the end.

Of course, you all heard the accounts on this or seen the movie and they chose payload without knowing her really well. But one of the selling points for her was that, okay, you’re not going to have a problem with the party base on this one. She’s a solid conservative.

Now when McCain picked Palin. I think that was a good signal in a way that Lieberman would have not been, would have been the opposite on this. Okay. For those conservative Republicans who are skeptical of McCain, or you could say a similar thing for Romney in 2012 or even Trump in 2016.

The base of the party might be looking at that person saying, who is this guy? Is he really one of us? One the most powerful ideas in politics, right? Is he really one of us? Is he a true Republican, a true conservative? If any of those candidates had picked someone who was liberal to moderate as a running mate, it would have been a signal to the rest of the party that, “See I told you he was a liberal.” Or probably moderate. “I told you McCain was a moderate.” Or, “For that matter, it shows that whatever he is, when he’s in office, he can take us for granted. He thinks he can win this thing without us.”

It might be how some Democrats feel progressive Democrats about Hillary Clinton picking Tim Kaine in 2016. On the other hand, when those candidates picked a conservative running mate, it’d be pretty hard to look at them and say, “Oh John McCain, he’s not a conservative at all. He’s so moderate even liberal.” I think how it struck a lot of people was instead to say, “McCain must not be as liberal as I thought he was.” Or moderate or whatever. “He must be more conservative than I thought he was.”

So take as in Biden’s case, if he picks, he has this somewhat moderate reputation. Depends on how you define that, right? But he’s kind of in the meeting of the Democratic Party, let’s say, pretty consistently as been throughout his career. Some folks, Bernie Sanders supporters, he’s too moderate to establishment. If he picks someone who fits that same mold, if he picks, let’s say Amy Klobuchar, for those people, it might just confirm for them, again, much like it did for Clinton picking Kaine in 2016.

“See I told you that he’s taking our boats for granted. He thinks that we have to vote for him because the opponent is Donald Trump.” Again, much like 2016 with Clinton. As opposed to if he picks someone who’s more progressive, it either shows that, “Oh. Hey, maybe Biden is not as modern as I thought he was. He’s more open to progressive ideas.” Or at least if he gets into office, he knows that he needs our support in order to survive and even thrive politically.

I think that’s the message that it would send and that’s based on our research. Why I think ideology matters, and I should mention we do in fact show, there’s limited evidence available for this, but we show in a couple of races, perceptions of ideology, the running mates ideology, influence perceptions to the presidential candidates ideology. That’s something we establish in the book in a way that really hasn’t been done in the literature.

Matt Grossmann: Adler says the parties do seek public and party input, but not necessarily based on perceived electoral advantages.

William Adler: Clearly the parties are interested in the views of the public. We found substantial evidence of that in various forms. Taking polls or asking for input in the case of Ford’s nomination in 76, asking people donating money to say, who would you want the president’s running mate to be? And just they kept all of these notes that people sent in. So they have their finger on the pulse of it, of who people might want.

At the same time, it’s not clear that they’re thinking about it in terms of, well, if we choose somebody from the Midwest that will help us win key Midwestern States. It’s not obvious that that’s a consideration that’s at the top of their priority list. And it’s funny because there really is this huge disconnect between what people talk about during the Veepstakes publicly.

Yeah, we need somebody from this region of the country because it will help us win the Midwestern States that we need to win this election versus internally within the discussions that are happening within the campaigns within the parties are not centered around that at nearly as much. There’s some discussion of it, but it doesn’t seem to be at the top of the list of things that are really animating the conversation.

Another example, which isn’t in this paper but as in work that is still in progress in draft form and we did some work on Lyndon Johnson selection of Hubert Humphrey in 1964. And one of the things that is very apparent from that, is that while Johnson of course thought about it in electoral terms, it never once occurred to him that he needed a Northern running mate necessarily because he was from the South.

He didn’t talk about it as a geographic balancing effort or an ideological balancing effort at all. It was much more shaped by the politics of civil rights, by the politics of the things that were going on in the middle of 1964. And so a lot of those things that the public is interested in don’t necessarily matter from the perspective of the candidate. There is also a turn, as I mentioned, towards will this person help us govern? Of course, that doesn’t always hold true. You can think about the obvious, counter example would be the choice of Sarah Palin in 2008 where the governing considerations were clearly not on the table there.

And that was clearly clearly not what they were worried about when that choice happened. But for the most part, that turn has had a lot of impacts on it. And thinking about it in terms of how can we represent what the entire party wants, seems to be a much more substantial factor.

Matt Grossmann: And Devine says qualifications, maybe the best approach anyway, 2008 shows the real effects of running mates on perceptions, but mainly by showing the cost of the unexperienced Sarah Palin.

Christopher Devine: That perception of Palin’s readiness to be president influenced perceptions of McCain’s judgment and also whether he was too old to be president. Likewise for Obama picking, perceptions of Biden’s readiness to be president influenced perceptions of Obama’s judgment and also whether he’s too young to be president. Now people really thought McCain was going to die in office, or are people now thinking that Joe Biden’s not going to survive his first term? I mean, yeah, more of an issue than it normally would be, but I don’t think that’s it so much.

Or even that someone who’s less experienced like Obama in way would come in having no idea what he was doing and being total amateur hour, I don’t think that’s what people were saying as much, but they were some reassurances it wasn’t going to be an issue. And if you have any concerns about McCain’s health and then he picked someone who sets off alarm bells about whether this person could take over.

I think part of what it shows, you can say conversely for Obama and concerns about his experience, when they pick someone who just exacerbates that concern. I think what it says to voters is he just doesn’t get it. McCain, you have these concerns about his age and his health and then he picks someone with questionable credentials clearly just to try to win the election.

I don’t think there is any question that McCain just surveyed the field and Sarah Palin was the most credible vice president. He was doing this to try to win the election. I think people see through that. I think they would see through that if Joe Biden did it in 2020, and it just caused them to sour on McCain in many ways just have a more general anxiety about his campaign, let’s say.

Now conversely, when Barack Obama picks Joe Biden, I think that, what that signals to people who are concerned about being a freshman Senator and having very little foreign policy experience and so on. Although he had some, I think what it says to them is this guy gets it. He knows that he’s going to need help in office. He’s competent enough to win the nomination. But yeah, it would help to have people who really have the experience in office advising him. And not just that you pick a qualified vice president, but then we can infer from that.

And that’s what often these picks are. Are going to the indirect effects, it’s a signal or a window into their future presidency. It’s a good signal that, okay, when Obama gets into office, if he’s elected, he’s going to pick someone who’s also qualified to be the secretary of defense, and secretary of state and so on. This is someone who knows his limitations and is going to build that into the way he runs his administration. I think those are the inferences that people are making from the vice-presidential picks.

Matt Grossmann: Looking ahead, Adler sees the Biden list as largely a list of current party stars.

William Adler: It looks to me like a list that any person in the party might have put together. Any other person running might have considered some of these names as well. You can imagine in a counterfactual world, if Bernie Sanders had turned out to be the Democratic Party nominee that a couple of these people wouldn’t be on his list because he would want somebody a little more to his ideological liking perhaps. But again, some of that is representative of the party. If you are the nominee, and that’s where the center of the gravity of the party is.

Maybe it’s okay to pick somebody who’s like you because that’s what the party wants. I’m thinking about say Bill Clinton’s choice of Gore in 1992 kind of a replica of himself, a moderate Democrat from the South. And maybe that’s okay under some circumstances, but you got to think about it in terms of what somebody in the party would want. So somebody like Kamala Harris or Elizabeth Warren, these are names that I think would have been on just about anybody’s list. Who was considering running mates in this cycle?

Matt Grossmann: Devine is expecting a cautious experience pick.

Christopher Devine: Most of the vice presidents in recent decades really, I’ve been pretty well qualified. Whatever you think of their politics. I don’t think Mike Pence was a stretch for the office. I don’t think Tim Kaine was, Joe Biden, Dick Cheney, Al Gore, all pretty well qualified vice-presidents. And I think if anyone’s going to understand that lesson, it’s going to be Joe Biden having held that office and having a real appreciation for the good that our vice president can do within an administration.

I’d be really surprised. I wouldn’t rule it out, but I’d be really surprised if Biden really made some kind of desperate gamble with his pick to win votes. And if he did based on research, I think voters would see through that and it would not pay off for him. I think it would actually undermine what is probably his greatest strength, which is perceptions of his experience, his competence, his qualifications for leadership.

Matt Grossmann: Adler says parties may learn from their prior picks and selecting the next ones.

William Adler: You do see some of that developmental factor, I think in terms of especially posts, a pick like Sarah Palin where people are a little more trigger shy to maybe choose somebody a little different and that may lead you to some more standard oriented picks. Think about Trump’s choice of Mike Pence for example. Pence is very much your standard issue. Conservative governor from the Midwest, white guy pick just the sort of fill in the blank choice in some ways that year.

And Trump was very publicly considering other people like Michael Flynn who became his National Security Advisor or Newt Gingrich or people who were really going to be out of the box in some way. And the fact that he landed in the box is probably a testament to the perceived need to, if you’re going to have a presidential nominee who is this different, you want a running mate was a little more standard and boring and down the center a little bit.

But there’s going to be a lot of idiosyncrasies here because ultimately it is one person making the call about who they want. And so there’s a lot of, in the background of it, what does the party stand for and what does the party want? But if that person at the top chooses something different it’s ultimately going to be their call.

But even in the case of something like McCain choosing Palin, you see he wanted somebody who would be exciting and different. But one of the things he ends up landing on is somebody who is very conservative and very acceptable to the conservative base of the party. The way Republicans generally do.

Matt Grossmann: The next step for Devine is looking for possible effects in 2020, especially based on race and gender.

Christopher Devine: Yeah, I think we’ll continue to write some pieces on this looking at 2020, probably in the future. What effect did it have? I think this will be a really interesting, not just because it’d be a woman running mate, another chance to test that, but also if you were to pick CCA rooms Kamala Harris or Catherine Cortez-masto or governor Grisham from Mexico. Whatever other groups are involved here. That could be a really interesting piece to study there’s some more evidence to work with. But to be honest, pretty satisfied with where this ended up. And I don’t see another part two of black lives matter but we’ll see.

Matt Grossmann: And Adler and Azari will be looking at the effects of running mates on the future of the party.

William Adler: In this case it seems Biden is very explicitly making it about that about what he thinks the future of the party is going to be. And so I would anticipate that this pic definitely will reflect that. One of the things we’re looking into in work that’s in progress is thinking more about, if we flip that causal arrow the other way, to what extent does the choice of a running mate effect the party going forward? I think it does to some extent. You think about Palin representing that populist conservative turn within the Republican Party, it’s kind of hard to imagine getting to the nomination of Donald Trump in 2016 without it.

It leads to certain developments that make that sort of populous more broadly acceptable, more exciting to the party, more willing to turn in that direction. Or the choice of Geraldine Ferraro in 84 maybe leading to an increased concern for diversity within the Democratic Party. It’s both reflective of what the party has become and is a signal of what it might be going forward. Jack Kemp in 1996 representing a very conservative turn, but also if you think about him as the sort of four runner to George W. Bush’s compassionate conservatism. Kemp was that guy before George W. Bush was, and it sort of then works in both directions, but like I said, this is something we’re still digging into a little more.

Matt Grossmann: There’s a lot more to learn. The science of politics is available biweekly from the Niskanen Center. I’m your host, Matt Grossmann. Thanks to Christopher Devine and William Adler for joining me. Please check out, Do Running Mates Matter? And The Party Decides (Who the Vice President Will Be) and then listen in next time.