Billions of dollars in donations will flow to candidates this year. Citizens suspect all that money buys the donors’ influence. But just how different are donors’ views in each party from those of citizens? Neil Malhotra finds that Republican donors are more conservative than Republican citizens on economic issues but Democratic donors are more liberal on social issues. Both parties’ donors are more pro-globalization than their voters. So which do the candidates follow: the donors or the voters? Jordan Kujala finds that donors make candidates more inconsistent with their electorates and increase polarization in both parties.

Niskanen Center – The Science of Politics · How Donor Opinion Distorts American Parties

Guests: Neil Malhotra, Stanford University; Jordan Kujala, UC Davis

Studies:What Do Partisan Donors Want?” and “Donors, Primary Elections, and Polarization in the United States” 


Matt Grossmann: How Donor Opinion Distorts American Parties, this week on the Science of Politics. From the Niskanen Center, I’m Matt Grossmann.

U.S. candidates will raise and spend billions of dollars in campaign money this year, and citizens have long suspected that all that money buys the donors’ influence. But just how different are donors’ views in each party from those of citizens? And which do the candidates follow: the donors or the voters?

Today I talk to Jordan Kujala of UC Davis about his American Journal of Political Science article, “Donors, Primary Elections, and Polarization in the United States.” He finds that donors make candidates more inconsistent with their electorates and increase polarization, though Republican nominees are less responsive to their electorates.

I also talk to Neil Malhotra of Stanford University about his Public Opinion quarterly article with David Broockman, “What Do Partisan Donors Want?” They find that Republican donors are more conservative than Republican citizens on economic issues, but Democratic donors are more liberal on social issues. Both parties’ donors are more pro-globalization than their voters.

Kujala says scholars have had surprising trouble finding donor influence, but we should be looking for changes in the kinds of candidates who win, not votes bought in Congress.

Jordan Kujala: Most Americans would think that donors have a large influence on the actions of members of Congress, pretty much because of the campaign contributions that they’re able to give. However, political scientists have been trying to search for the influence of money in politics and on policies and votes in Congress for decades now with really little success.

There’s been really little evidence that donors’ money in politics has affected the levels of polarization that we’re seeing in America today, with Democrats and Republicans in Congress being … They’re not really agreeing with each other on really any issues.

To actually hit this point home just a little bit, about a week before this paper was accepted by the American Journal of Political Science, there was a good article on by Maggie Koerth … I don’t know if I’m butchering that name or not; I apologize if I am … called Everyone Knows Money Influences Politics, Except Political Scientists. It lays this out a bit of what the general public thinks versus what political scientists do.

Of course, this isn’t to say that political scientists have found no evidence for donor influence. For instance, donors appear to have easier time getting in-person meetings with members of Congress and their staff, but these findings have tended to be limited, and the overall influence found is not necessarily that large. Definitely less influence than most Americans would think.

These findings have even led some political scientists to suggest that money has little or no real effect in public policy. And so, I argue with that conventional wisdom basically, that money has a limited influence on public policy and basically votes in Congress.

Basically, many scholars have tried to analyze the direct relationship between donations and money outcomes. But I don’t think that’s really the appropriate way to look at that relationship between money and the ones being elected and what their policies are.

Just given what we know, that’s not really how money influences. You’re not going to be able to see an elected official get this huge sum of money and you’re going to be able to tell that official change their votes in Congress because they got that sum of money then. It just really isn’t going to happen that way.

Instead, donors are able to influence folks in Congress by getting people that they agree with elected in the first place. So [inaudible 00:03:21] we need to analyze that relationship within the congressional district, at the district level, between the donors and who actually gets nominated from those parties to see what influence donors have actually on who gets elected then.

Of course, this was actually the main difficulty in previous works, was that it was really hard to get a large sample of donors and a large sample of candidates together in order to do this, because if you want enough sample of donors and candidates, you need 50, 100, 200 candidates, I think 3,000, 4,000, and you need at least a certain amount of people per district then, donors, right? So you need to know, if you want 50 donors [inaudible 00:03:55] 50 times, 3,000 to get what you may want there.

I do find evidence that not only does money influence politics, but that basically the policy preferences, the ideology of congressional candidates, and the roll call votes of members of Congress are actually most responsive to the donors in their district. So we’re almost going back to the conventional wisdom of Americans where money does appear to be quite influential in politics, but maybe not just in the way that Americans think it is.

Matt Grossmann: Malhotra says their findings fit a conventional story about each party’s elites, even if it doesn’t fully support donor extremism.

Neil Malhotra: I think there is this general conventional story, which is that donors are more extreme and they’re producing a lot of polarization. That’s also akin to some of the previous findings on donors in the political science literature. But I would say that there’s a secondary conventional wisdom that some academic papers have touched on, not with donors but with the wealthy, which is more that you have to look at the specific issues.

So on many of Fiorina’s famous books on polarization, he speculates to this point without a lot of hard data. But based on the point that Fiorina’s making is that when you look at these kinds of either high-level donor dinners or even when congressmen are dialing for dollars, the donor class they’re in touch with, on the Republican side, they’re just very concerned with economic issues like tax cuts and regulations. On the Democratic side, they’re not bringing up new deal issues related to labor, but they’re bringing up post-materialist things about the environment and social issues, abortion, gay marriage, et cetera.

So that’s the Fiorina argument on why the Democratic Party’s become a little bit more extreme on the social issues and gotten away from their new deal roots and why the Republican Party seems to be forwarding a lot of economic policies, which are out of step with the median voter, not only generally but in their own party.

So I think it matches this secondary conventional wisdom, but it provides more nuance to the basic finding that, oh, donors are more extreme because maybe they care more about politics, et cetera.

Matt Grossmann: Malhotra’s project built on a larger study of changes in elites.

Neil Malhotra: David and I have, I think, been generally interested in this topic of the diversion policy preferences of masses and elites and how it may speak to the rise of populism and maybe threats to the democratic order.

I think this is a big question these days, especially as elites start gaining more power in the economy and, as a result, potentially more political power as well. So you can go back to Charles Lindblom’s work and [Pershonen 00:06:49], and there’s just a lot of arguments that once you have economic power, there’s a lot of reasons that would give you political power as well.

And so, one group we were just very interested in studying was Silicon Valley elites, mainly because a lot of the wealth creation has been in that sector. If you just look at the top companies by market capitalization in the US, I believe the majority of them are technology companies now, whereas I think it used to be oil and gas and finance probably 10, 20 years ago.

As part of that project, we also collected just as a baseline data on just donors, because I think people would be interested in knowing is this something specific to technology elites or is it also specific just to rich people generally? Then we decided that was a very extensive data collection effort. And so, we wanted to delve more into the donors specifically, including seeing could we replicate it in other datasets that has analyzed donors as well?

But I think this project is part of a number of research on how elite attitudes differ from mass attitudes and what that says about our politics, and more about whether, I think, this general neoliberal consensus, which has dominated American politics since World War II, has some fragile basis to it.

Matt Grossmann: And Kujala has been looking for a long time at money in politics, trying to find the path of influence.

Jordan Kujala: The core of my research is looking at representation of elected officials in America. How well do elected officials actually represent their constituents in their states or in their districts that they come from? I’ve always been interested in representation and the influence of money.

Actually, my first foray into analyzing this influence came back in undergrad, when I was an undergrad at Grinnell College in Iowa, looking at the relationship between PAC contributions and vote-share in the general election.

I was basically taking a political parties class with my good adviser Barb Trish. We were required to do this quantitative analysis for class. I’m like, “Let me look at the influence of money in politics.” I come from a working class background. It seems to me money influences everything. So, of course, money influences politics, right?

I was wrong. I looked at that relationship between PAC contributions and the share of votes in general election, and I found absolutely no relationship. So after that, I really started to learn and get into this more about how little evidence there is in political science research for the influence of money in politics.

I ended up going to the University of California Davis to get my PhD with the intent on analyzing that influence of money in politics. But again I realized that there really wasn’t a lot of appropriate data to analyze it in the proper context that I think it should be.

Luckily for me, Adam Bonica at Stanford created the DIME dataset, that was the first to produce these ideological scores from large numbers of donors. And so, with his dataset, I’ve actually reworked my dissertation.

So it’s been a long process, this paper has been. This was the main part of my dissertation then. And so, that whole process has taken years basically to finally look at that influence of donors on policy preferences.

Matt Grossmann: Overall, Kujala’s findings show that candidates are responsive to donors, and that helps polarize American politics.

Jordan Kujala: In my paper, I examined the influence of donors on the ideological polarization of Republicans, Democratic nominees for the House of Representatives. What I find is perhaps the strongest evidence to date that the influence of donors in primary elections is the source of political polarization in the United States.

These results suggest that Republicans and Democratic nominees for Congress, specifically the US House, are more responsive ideologically to the partisan donors in their district than they are to their primary or general electorates.

As donors take more ideologically extreme positions, primaries’ winners take more extreme positions that are actually further from their district. Ideologic Democrats appear to be most responsive to their donors and primary constituencies while Republicans appear to be only responsive to their donor constituencies. Overall, I find that the polarizing effects of donor constituencies tend to dominate any moderating effects of elections, leading to extreme nominees and ultimately members of Congress.

I think probably the most important implication … And there are many, I think, potential implications from these words. But I think the most important is probably that these findings provide evidence that affluent Americans may be able to actually use their wealth to influence political outcomes. That means the [inaudible 00:11:09] inordinately respond to their party’s donor base, which is a group that is disproportionately wealthy, which may ultimately lead to policy outcomes that favor the wealthy over the middle class and the poor.

Matt Grossmann: He had to compare candidates to donors, primary voters and citizens on the same ideological spectrum.

Jordan Kujala: The real main contributions of my paper is the ability to kind of construct a dataset that actually contains these ideological measures for house candidates, as well as these important district level constituencies, such as donors and primary constituencies in general electorates, just because it’s been difficult to do that in the past.

What I’ve been able to do that is I have about 3,000, 4,000 candidates and then I’m able to actually then get their ideological scores, and I’m actually able to then take ideological scores from all these other surveys. No dataset has all of these things together. I have to basically combine a dozen different datasets and data things from different years to actually make these things.

I’ll actually take the donor information and the candidate’s ideology from [inaudible 00:12:09] datasets. I then take constituency information about the general electorate and the primary constituency from the Cooperative Congressional Election Study, and then I use other datasets that’s from my adviser Will Stone that actually allows me to place these things together onto the same scale, which is really important for understanding this because most of our ideas and theories behind what actually is affecting policy preference of members of Congress requires us to actually have these measures on the scale to actually properly kind of test them. So doing this and being able to bring them all together was really a key contribution, I think.

Matt Grossmann: He found donors were a more extreme constituency, but there are differences by party.

Jordan Kujala: Across party, you tend to see that donors are more extreme than the average. Let’s say for Republicans, the average Republican donor is more extreme than the average Republican and the average Democratic donor is more extreme than the average Democrat, although there’s a lot more consistency or similarity between Republican partisan constituencies and Democratic ones. And in general donors tend to be more extreme than members of Congress. So what I tend to find is that incumbents are the most moderate, open seat candidates are next, followed by all party challengers and that, really, it seems to be that donors are kind of more extreme constituency that does maybe affecting kind of what’s going on.

There’s a big difference between Republicans and Democrats in terms of that relationship between donors and primary constituencies. In every single district, the Democratic donor constituency in my measure is more extreme than the Democratic primary constituency. For Republicans, it’s a lot more mixed. Only about 60% of Republican donor constituencies are actually more extreme than the primary constituency and that the average Republican donor constituency is really, really close on the seven-point scale to the average Republican partisan constituency. You still see that donors are more extreme for Republicans, but you see a lot more overlap and similarity in terms of their policy preferences with the rank and file members of their party than you do for Democrats.

Matt Grossmann: Malhotra and Broockman focused on party differences, comparing donors with voters in each party on what they believe by issue area.

Neil Malhotra: The main goal of the paper is to produce descriptive statistics, comparing the policy views of donors and the mass public. So the goal is to try to get both groups of people answering the same questions at the same time so kind of we can compare them and make some descriptive inferences that might help people who are doing their own research on the topic, and then people in the general world.

I think there have been surveys of donors before, but we wanted to focus on two new aspects of the survey. First is we wanted to specifically oversample and target very, very large donors, including people who donate tens of thousands of dollars, because we view them as qualitatively different than people who barely donated enough to show up in FEC data. Moreover, we wanted to ask questions that got at various dimensions of policy issues, including economic issues, social issues, and issues related to globalization, to look at heterogeneity by party and policy domain. I think by doing these two things, we produced a lot of new findings that have touched on things people have known, but we’ve never really were able to compare donors and voters directly.

The first thing we found is that it is true that, overall, donors were more extreme in their policy preferences than the mass public, but there actually is a lot of heterogeneity depending on what party you’re looking at and what policy domains. For economic issues, Democratic donors are about the same as Democratic voters, whereas Republican donors are much more economically conservative than their voters. The pattern is actually flipped for social issues, which is that there is a lot of similarity between donors and voters on social issues, but on economic issues, Republican donors are much more economically conservative and extreme. On the globalization domain, both groups of donors are more pro-globalism than their counterparts in the mass public.

So although we don’t really have evidence of causality in this paper, it does kind of point a picture, which a lot of people have hinted at, which is that the donor class is sort of moving the Republican party more economically extreme, it’s moving the Democratic party potentially more socially extreme and is creating this chasm between masses and elites on issues related to globalization.

Matt Grossmann: For example, universal healthcare attitudes and same-sex marriage show the patterns of differences by party.

Neil Malhotra: I’ll give you an example of a few issues, which I think kind of show the pattern. We asked about a lot of issues and not every issue matches this pattern, and we kind of transparently explain that in the paper. But these are just, I think, some typical issues, which reflect the pattern overall.

For example, we asked about universal healthcare and 52% of Republican donors strongly disagree that the government should make sure that all Americans have health insurance and only 23% of Republican citizens felt the same way. So that gap of 52 and 23, about 30 percentage points, is humongous and really kind of does show that even a lot of Republicans offer a lot of liberal attitudes on economic issues related to redistribution and the donor class in the Republican party just has very different views. I think they’re much more conservative economically, more similar to what you would call like economic libertarianism.

On the other hand, if you look at the social issues, they tend to be pretty similar, so I’ll give one example of that. For example, if you look at same-sex marriage, the Republican citizens who strongly oppose same-sex marriage is about 30 percentage points, and that’s very similar percentage points in the Republican donors. If you just look at the distributions on that issue, they’re almost identical with a slight majority of Republican donors and citizens opposing same-sex marriage.

On the other hand, if you look at the Democratic party on very similar issues, you find that, if anything, the donor class is a little bit more liberal on economic issues. If you look at the health insurance question that I’ve mentioned earlier, Democratic donors are actually way more likely to support universal healthcare with over 75% strongly favoring it, whereas in Democratic citizens, that number is a little bit under 50 percentage points. But on the social issues, for example, same-sex marriage, the Democratic donor class is much more extreme. They get almost universal support among the donors for same-sex marriage and among Democratic citizens only about a little bit more than 50% strongly support it.

Neil Malhotra: Those are just some examples of where you have kind of divergence across party and issue domain in the gap between donors and voters.

Matt Grossmann: But elite mass divides on globalization appear in both parties.

Neil Malhotra: On the globalization items, this pattern is much stronger for the Democratic sample than the Republican sample. Part of that is, I think, just the Republican party as a whole has moved more antiglobalization over time, but you still see the results. It’s just that they’re much more dramatic for the Democratic group.

But one kind of common question that’s asked in public opinion research is should we pay less attention to problems overseas and concentrate problems at home? Over 30% of Democratic citizens strongly agree with this, whereas among Democratic donors, that number is between 5% and 10%, so that’s a pretty big gap. Then you see something similar among Republican donors, which is that about 15% of Republican voters say that we should pay less attention to problems overseas. And that same percentage points, or the same figure is 25% among Republican donors. That’s very similar 25 percentage point gaps between both donors and voters in both parties, even though, overall, Democrats are more supportive of globalization.

Matt Grossmann: Extremism just means big divides here, not necessarily far left or far right views.

Neil Malhotra: We’re not trying to say objectively or normatively what’s an extreme opinion, but more just comparing what the beliefs are. I mean, if you don’t want to use same-sex marriage as an example, the death penalty is also a very nice example where there’s maybe more division in the public. But the majority of Democratic citizens are in support of the death penalty, whereas the majority of Democratic donors are against the death penalty, whereas among Republican citizens and donors both basically have identical strong support for the death penalty.

I guess our paper doesn’t really delve into the question of what the true fixed point of extreme or not extreme is relative to the status quo, but just kind of what the gap between the two groups are.

Matt Grossmann: They matched their results from regular surveys that include donors.

Neil Malhotra: Greg Huber and Seth Hill had written a paper, which is not part of the regular CCES, but it was something they did specially, which was to link the CCES, which has just such a high number of voters to the FEC records. And so they’re figuring, “Oh, well we’re doing a large scale survey anyway. So why not just link to the FEC records? And then we can compare donors and non-donors.” So when you do that, and you’re not specifically trying to sample donors, like we did, you’re just going to have a… The typical donor is going to donate a lot less money, which is fine, but this gives us a great opportunity to see if our findings were idiosyncratic to our sampling strategy or our questions. So we did a preregistered replication where we said, “These are the questions that are going to be assigned to these three domains. And then we’re going to run the analyses, including what the code would be on the Huber and Hill dataset.”

And so we did that, and, generally, the findings matched up. Now, if you look at the Huber and Hill dataset, there’s a lot of differences. So their questions that they’re asking about are very different than ours. So their globalization items were read a lot about military intervention because that’s what the CCS was asking about at the time. And ours are more general or about free trade and immigration. But I think I do that as an advantage, which is if you had the exact same study and the exact same questions that you were trying to replicate, it would get to the narrow replication point, but it wouldn’t get to the conceptual replication, which is this general finding robust to different types of questions in different contexts and the different way researchers approach this question. And we found that it was.

Matt Grossmann: And they found that the top 1% have even more differentiated views following the patterns.

Neil Malhotra: Everything I mentioned is, basically, stronger among the top 1% donors. So all the patterns about heterogeneity by policy and party just become stronger and affect size and statistical significance among this smaller group of top 1% donors. And so after we weight the data that is going to be half our population. So even though it’s top 1%, they’re actually half of our data, because it was a specific over sample. So I think that shows the findings even more. Whereas I don’t think we’re trying to make any causal statement about the amount you donate and your extremity, but descriptively, it does show that if we believe the top 1% are more influential than the top 99% just based on the amount and their involvement, ll the patterns and discrepancies in views that I’ve mentioned are even more heightened.

And then I think this does tie to research about just generally wealthy people. So Elizabeth Rigby and Cory [McSolomon 00:25:12] published a paper very similar time period as ours doing similar analyses, but just using survey data on income and looking at wealthy or high income people, low income people. And they found a very similar pattern of results, where you, basically, see that in the Democratic party, the masses are much less socially liberal. And in the Republican party, the masses are much less economically conservative.

Matt Grossmann: Kujala tried to assess causality, finding that primaries increase the importance of donors.

Jordan Kujala: While electoral success may ultimately depend on voters, even if you get enough votes to win an election, you can’t actually win that election unless you have enough resources to run a campaign. Elections at all levels are costing more and more money. And only those who are able to generate enough of that money are actually able to have any chance of winning. So, really, this puts donors in an advantageous position during the primary to constrain the policy of those that are potentially nominated. As Kathleen [Bondadall 00:26:10], in their 2012 work note, because this is a theory of political parties, groups, policy demands, and nominations in American parties… And sorry in American politics, contributors are particularly advantaged during the primary because candidates must compete amongst members in their own party for resources. There’s no party label to help you to generate who you’re going to give money to.

And the resources necessary to actually win a primary are often small relative to the general election anyway. And so I would actually say that many districts, the vast majority, there really are no trade offs for these parties. And in many districts, you win your primary, you’re going to win the general election regardless of how extreme you are ideologically. So in many [inaudible 00:26:46] districts, say, for one party, donors can potentially actually try to demand as many policy concessions as possible. There really aren’t a lot of competitive districts. And in those districts, it can become an issue, but overall, it really seems that they’re able to kind of make these demands or at least potentially make these demands without losing. Although, we do find there’s research by Andrew [Hawley 00:27:06] who find that when you do pick a motion candidate in a primary election, it can lead to the other party winning the election. But I think in most cases, it’s unlikely to make a difference in the election.

Matt Grossmann: And he may have found a reason for the effect. Donors do veer away from candidates who aren’t close to them ideologically.

Jordan Kujala: What I’m trying to do is… What’s going to determine if the relationship if I find… if that’s [inaudible 00:27:26] relationship is based on the actual influence of donors or what we refer to as lawfully, right? That is could it just be that the nominees are not influenced by donors. They just kind of resemble them. Because nominees and members of Congress almost always live in their congressional district, and they have higher levels of education income that are similar to donors. And, actually know many of them are donors themselves. And so, as a result, it’s possible that I’m not finding actual influence but resemblance.

And so this is kind of getting at a first cut at just seeing is there even conditions present for influence, right? At the district level, are donors actually giving money based on how close you are ideologically? And I find that it does seem to be the case. And so donors may actually be done demanding these policy concessions in exchange for resources. But then again, this is just suggesting the conditions are present. It doesn’t actually mean that we conclude that this is the relationship we’re seeing is because of this influence. And so I think this is kind of a future step that needs to be done and works just to parse out that relationship and that influence.

Matt Grossmann: But he acknowledges it’s hard to see which came first, given that ideology is measured through campaign donations.

Jordan Kujala: So the first thing I try to do to bypass that link between house nominees and house donors is I only use donors at the district level that contributed to a presidential campaign, which I think is a good idea, but I don’t know if that was too persuasive. I think what was persuasive was that I’m sure my findings weren’t solely based on my choice of candidate ideology using the [Bonica 00:28:50] data, using the [inaudible 00:28:52] data, was to actually use different measures of ideological scores for candidates, put those onto the same scale using the same methods that I used before, and then test those relationships.

So, in particular, I replicated my findings using probably the most common measure professional ideology, DW-NOMINATE, which is based on roll call votes in Congress. This can be found in the appendix of my paper. And then, when I run these analyses using those measures, the findings are subsequently similar. I find that both the members of Congress and the house representatives are more responsive to their partisan donor constituencies than to their primary or their general electorates. And there’s no circularity issues with the DW-NOMINATE measure like there is with the Bonica measure.

Matt Grossmann: Now, Malhotra says their findings fit with Kujala’s paper, but he says we need to look for issue differences.

Neil Malhotra: I think that paper is trying to get at the causality more. And I think causality is just very hard in this specific research question. But looking at over time responsiveness is one way to do it. So I think nothing we’re seeing in our paper is inconsistent with that one. And I would just be interested to look at a lot more of the heterogeneity by party and issue domain. Because I think one hypothesis that our descriptive data would suggest is that the Republican response in this issues would be much stronger on the economic dimension. So we’ll see if that’s true if you look at that data. If you don’t condition by issue domain and most of the issues you’re looking at are economic issues, then you could produce results that look like this. So I would just wanted to point that out. I think the issue of heterogeneity is important to just test for.

Matt Grossmann: And Kujala says the donors may be changing, but it’s hard to see who is following who.

Jordan Kujala: The story goes similarly, right, where we do really see that no conservative donors, and I think you’re going to talk to Dave Brockman about this too, where conservative donors are more… Sorry. Republican doors are more conservative on economic issues and Democratic donors are more conservative on social issues. And we’ve seen that shift over the past 20, 30 years, right, where economics has seemed to shift to more conservative positions in terms of what we were talking about and what is the debate. And socialists have shifted more liberal in that way. It’s hard to tell if there’s actually a change in donors in this, right?

I mean, all of my data comes from 2002 to 2010. I don’t have anything before or after right now. And it’s unclear. I’d be really interested to see how the composition of donors has changed over time. But it’s just really difficult to do. And so I imagine there’s probably been some changes in that way. And I think there’s probably got to be some sort of feedback loop in here, somewhat, where donors are changing some and then to get money, then they’re going out. Candidates are the ones that are able to actually come out and actually run for office that are kind of changing, which then changed what donors maybe are contributing to. And so it’s really difficult to say how those changes actually affect what’s going on.

Matt Grossmann: He says politicians are still reflecting their donors.

Jordan Kujala: Once again, this is 2002 to 2010. So one of the things that I’m not really able to account for is the Citizens United ruling time, more outside groups and their spending on this, which is something I want to look at in the future, is how has that affected [inaudible 00:32:00] relationship or donors potentially. But I mean, I think you are seeing that a little bit. I mean, you’ve seen that with some of these primary challenges, maybe you’re seeing with in Illinois, Representative Alexandria Ocasio-Cortez, although, GW-NOMINATE paints her as more moderate member than [inaudible 00:32:14] do. But I do think you’ll kind of see a little bit more of that, in part, because of just the fact that you’re seeing… I mean, we’re seeing right now, right, during the COVID-19 pandemic, where’s the money going from members of Congress? Where do they seem to be most concerned about in terms of helping the economy?

And it’s not necessarily average Americans. And I think you can see from there that you’re seeing a disconnect between members of both parties, in some [inaudible 00:32:40] and the overall electorate itself in terms of what they’re looking for. So I do think that we should still see some of this, where I would suggest that, if I was a Democratic operative, trying to basically create the most… have the most Democratic, most partisan, most ideological, liberal people in my caucus when I’m in the House of Representatives, is I basically take any safe district that Democrats had and I’d try to to run big, as much as a lot of the districts they hold.

And you could run basically anybody of almost any ideological extremity in that district and they’re going to get elected in the general election, so long as they win the primary. Which I think is a little bit what we saw with the tea party and Republicans, which is they found that they’re more extreme. They’re able to kind of knock off some potential, more moderate Republicans. It doesn’t mean they’re necessarily going to lose the general election. They might, but in most districts, it’s not going to make any difference. And so, whether or not, I’m not sure what should happen, but if Democrats want to have the same power you guys, you Republicans do from that wing, which may not be the case, I would definitely suggest trying to do that. Although, that’s a whole different ball game then too.

Matt Grossmann: Malhotra says, “Future research should look at issue-specific trends and top donors.”

Neil Malhotra: Two big takeaways are, is that one I think looking at heterogeneity by party and policy is going to be very important for those scholars. Because if you just look at everything on a single issue dimension, you might be obscuring a lot of differences and you could be producing null or weak findings, whereas the results could be a lot stronger. Because keep in mind, if we had just not separated our results out by policy domain, which we did a priority, the results would just look a lot weaker, right? And so I think people who are studying donor influence should take that into account. And especially now we have multiple papers replicating this finding.

And then I think people should be looking at the top donors specifically because I think putting more casual donors in there does dilute these effects a bit. And especially as we get access to more casual donor data, small donors from Act Blue, things like that. But I think the general takeaway of the paper hopefully, will inform influence as people try to look at policy effects, which is, is the donor class an explanatory variable for why the Republican party seems to be out of step economically and potentially why the democratic party seems to be out of step socially with their own voter groups? So even though we don’t really have causal evidence, I hope that people who are doing those causal studies will look at our paper and take something away from it.

Matt Grossmann: Kujala says, “Donor effects are likely operating through limiting the pool of candidates more than vote-buying.

Jordan Kujala: Each individual candidate is not necessarily having this overall strategy based off of where donors are and where the candidates are, and they’re trying pinpoint their policy preferences based on that. I think a lot of this, I mean, I think a lot of what occurs in terms of who gets elected, who ends up running for office, who they listen to, has to do with these networks; has to do with who was even possibility to be a candidate for a House of Representatives at any district. Because there really aren’t that many candidates out there and so really, there’s a limited pool there. And so I think that these type… what really, that money is doing is it’s limiting the pool of candidates that could possibly get elected.

And so it’s not necessarily that these candidates are going out and specifically doing these things, it’s that these are the ones making contact, these are the ones going. And then they don’t realize they’re necessarily out of step with their districts because their networks are telling them that they’re not. And so I think this actually works really well together in terms of how they’re getting elected, why they’re getting elected and why they’re not… they don’t seem to be as responsive to these groups as they are. And so that seems like it could be one of the mechanisms to why they’re not as responsive is because they are resembling the donors and then they’re only hearing from people that are maybe donors or have similar policy preferences to donors.

Matt Grossmann: While Malhotra acknowledges that donors could just be one category of party elites with differences just indicating broader divides between the parties and their voters.

Neil Malhotra: I think we can’t speak to if this is the donations per se, because there’s a lot of features of donors that are similar to other activists. So with donors, their resource that they can expend on the margin is their money. And for other people, maybe the resource they can expend on the margin is their time. And you would have similar processes, like you really care about a few issues and are extreme about them.

Neil Malhotra:

And I think if you replicated this analysis among convention delegates or party activists, which I hope people do, I think you would find very similar things. So, I think we just substantively believe that donors are important because they fund campaigns in a system where raising money is very important. But if you look at, for example, convention delegates, and I wouldn’t even just talk about national convention delegates, but there’s many states where state-level convention delegates and local are very important. And maybe those people are not donating as much, but they can influence a lot by mobilizing voters and helping pick primary winners, things like that. And so I think if you found similar results it would, I think, speak generally to this issue of: who is most involved in American democracy and are their opinions distorted in any specific way?

Matt Grossmann: Next, Kuala wants to look at changes in outside group spending and what’s driving the candidates now.

Jordan Kujala: What’s happening now? I’ve got all this out there. I’m still pushing some things up in my dissertation, looking at one thing we get is party convention. So some states, Utah, Connecticut and some districts in Virginia don’t even hold primary elections. So I’m looking at what effect… other things that might affect the relationship between these constituencies and the policy preferences members of Congress choose.

But I think one of the things I want to do is, what’s happening now? What is happening since, with this newer district team between 2012 and 2018 elections? What is the influences of these outside groups? Does it affect the influence of individual donors in the district? Does it not affect it? Doesn’t augment it in some way? I think those are really important things to think about because obviously, there’s a whole lot of speculation and argument about the influence of the Citizens United ruling and what that might mean. And I think this is one way to really look at: what are the these differences? What changes might we see in these relationships that we haven’t really been able to look at too in depth before?

Matt Grossmann: And Malhotra is taking on a broader look at whether a long-running elite consensus is increasingly being challenged.

Neil Malhotra: I think this general topic of what the elite consensus, policy consensus is and how that might be being challenged or dismantled, I think is a very important topic. Because I think there’s been a lot of trust in elite consensus for many, many decades and it’s being challenged a lot in both parties, where you have a lot of things on the economic issues, which I think a lot of people would view as very extreme, but a lot of the survey data showing that many people actually agree with those positions. And also, in terms of how we’re viewing globalization in a default [predo 00:40:05]-superior thing that’s being challenged.

So I think Dave and I would love to in the future, look at more of these type of elites groups and try to get maybe the differences to see, is it donors specifically, is it business elite specifically? In our prior work, we found that Silicon Valley elites are actually very different than donors, that it’s not just that some people are rich and some people are poor. So I think more investigation into that, looking at a lot of the groups you’re talking about, like who are at these party conventions? What about people in other very strong in industries like finance? Things like that would be great to study.

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 Grossman. Please review our recent episodes at or anywhere you find your podcasts. Thanks to Jordan Quila and Neil Trachoma for joining me. Please check out Donors, Primary Elections and Polarization in the United States and What Do Partisan Donors Want? And then listen in next time.

Photo Credit: By Michael Vadon – Own work, CC BY-SA 4.0,