Trust in government is low and declining after another polarized election and in a polarizing administration. We’re frustrated, even angry, but maybe we still hold some underlying pride in our government. Steven Webster finds that anger decreases trust, whether it’s directed toward the candidates or life in general. But Stephen Nicholson says we still hold implicit trust in government that can be drawn upon in a crisis. We discuss the two sides of trust in government.

The Niskanen Center’s Political Research Digest features up-and-coming researchers delivering fresh insights on the big trends driving American politics today. Get beyond punditry to data-driven understanding of today’s Washington with host and political scientist Matt Grossmann. Each 15-minute episode covers two new cutting-edge studies and interviews two researchers.

You can subscribe to the Political Research Digest on iTunes here.


Grossmann: This week on Political Research Digest, why Americans implicitly trust government, but say we don’t. From the Niskanen Center, I’m Matt Grossmann.

Trust in government is low and declining after another polarized election and polarized administration. We’re frustrated, even angry, but maybe we still hold some underlying pride in our government.

A new study finds that anger decreases trust, whether it’s directed toward the candidates or at life in general. I talked to Steven Webster of Emory University, author of “Anger and Declining Trust in Government in the American Electorate,” published in Political Behavior, about his work on emotions and distrust.

But is there some remaining trust lurking under the surface? I talked to Stephen Nicholson of the University of California, Mercer, about new evidence on implicit trust. His new study, in the Journal of Politics, is called “My Trust in Government is Implicit, Automatic Trust in Government, and System Support.” It finds that although explicit trust is necessary for supporting particular policies, implicit trust can help with broader system level support and when the country is faced with crisis.

Scholars often see trust in government as, in large part, a product of partisan politics. But Steven Webster says it’s also about anger.

Webster: Conventional wisdom is that trust government is really shaped by partisanship. So if I’m a Republican, I’m going to have low trust in government when Democrats are in control and vice versa. And my study doesn’t refute that. I find that partisanship still plays a large role, but anger plays a pretty large rol as well even when controlling for an individual’s partisanship. These effects are almost on par with one another.

Grossmann: And anger has been rising over time.

Webster: There’s plenty of angry people in both Democratic and Republican parties, and so really what matters is how politicians can sort of shift people around their main levels of anger. And I think anger is increasing, specifically because the contemporary era is so polarized that it creates incentives for politicians to invoke anger and fear towards the opposing party. So I think anger is certainly increasingly used as a campaign tool.

Grossmann: Trust is also declining and the trends may go hand-in-hand.

Webster: The government has been declining a long time and that’s sort of one of the paradoxes that a lot of people have pointed to is that people tend to trust in their own member of Congress, but not Congress in general.

So I don’t really show anything different than that. I mean, you can look at trends from Gallup or any polling organization and you can see that trust has declined precipitously over the past 50 years. And so I don’t think I’m necessarily differing from anybody. What I’m showing is that there’s other factors that have been previously been overlooked. And I think this factor, anger, is so important now because the contemporary political environment is characterized by anger and negativity.

Grossmann: Webster sees trust intertwined with other attitudes toward government.

Webster: I think they’re very tightly wound up with one another. It’s hard to know: Does trust affect efficacy or is it the other way around? It’s sort of a classic chicken and egg problem, so to speak. And so I’m not, you know, willing to take a hard stand on what’s efficacy and what’s trust because I think they’re largely tapping into some sort of a way in structure that is similar. And so what I did in the paper is I would looked at various measures of trust or efficacy, if you want to call it that.

Webster: So I look at whether people believe the government is crooked, whether the government cares about ordinary people, whether ordinary people have a say in government. And then when I do my experimental study, I’ll look at whether people believe the government is responsive to the concerns and interests of ordinary people. So I really try to triangulate and hit all different angles of this measure of trust in government.

Grossmann: He measured anger towards the candidates in surveys, but also did an experiment stimulating anger.

Webster: The first measure I used for this sort of targeted form of political anger is the frequency in which individuals feel angry at the opposing party’s candidate. When I moved to this more broader notion of anger that’s generalized or a political anger, I look at the frequency of anger at either party’s candidates. And so I’m not convinced that these are perfect measures by any means, generally what we have to make do with. So that’s what I think the most convincing part of this study is the experimental part.

Grossmann: In the experiment, Webster had some citizens make themselves angry.

Webster: I asked people to write about various things. Some asked to write about a time they were angry about politics, some people were asked to write about a time they were angry, and some people were asked to write about a time they thought about politics. And the reason that I split these groups apart is because I wanted to separate the emotions, so anger, from the target, politics. And so what I found is that anger that is specifically about politics does reduce trust in government. People tended to use more emotional words, but this was also true for people who just read about a political thing. And so I think, really, anger in general, not necessarily politically motivated anger, is what’s driving this here.

Grossmann: Both types of anger made trust in government decline.

Webster: Anger can actually have a profound impact on altering trust in government. Specifically, what I found is that when people are angry, they tend to exhibit lower levels of trust in the national government.

Grossmann: Webster says professionals are learning from the research, but not to our benefit.

Webster: They’re learning the lessons, but those lessons aren’t good for democracies that large. So here is Georgia recently, we just had this special election between John Ossoff and Karen Handel. And the advertisements on TV here were negative, you know, John Ossoff was tied to liberal San Francisco values and Nancy Pelosi and every ad you saw was trying to evoke anger.

And I think political professionals know that anger works because the current era is one where we don’t really need to motivate people in order to vote for a candidate. We need to make them angry and fearful towards the opposing candidate and this fear and anger is what makes a loyal voter. So I think political professionals and politicians are learning from political science, but they’re learning things that aren’t necessarily good for society as a whole.

Grossmann: And he sees the negative consequences for political legitimacy continuing.

Webster: Without trust in government, it’s hard to imagine people seeing the representatives and the laws that we pass as legitimate. I especially see some of this now with the current administration, you see people on the political left saying Trump is not a legitimate president because he maybe colluded with Russia.

And so I think that has problems for sort of Democratic governors and representation. People do behave as if they trust the government. My concern is that this anger will continue to go so much that we really will start to see some serious consequences.

Grossmann: Yet Stephen Nicholson’s research says citizens may still have some underlying trust.

Nicholson: People posses this deep-seated, gut-level trust in government and this is a meaning or consequential attitude. It affects why people have this kind of overall support for the political system as well as those instances where there are times of danger in the United States. So, a national crisis or an international attack. We think that implicit trust is a meaningful attitude that’s solicited during those periods of time.

Grossmann: His work does not challenge the importance of the conventional measure of trust or its fall, but he sees hope in their new measure.

Nicholson: There’s a great deal of consensus about trust in government and that is that it is very low. Academics, media, pollsters are all in agreement that regardless of the type of questions asked, Americans express very low trust in government. In fact, often times, you’ll see only about 20 percent of the American public expressing some sort of trust in government, meaning 80 percent are distrustful.

And we asked the same sort of questions in our survey and they didn’t differ from what’s come before, so we haven’t disputed these previous studies, we don’t challenge this idea that people aren’t distrustful of government.

What we do challenge is that this is really the only way of thinking about it and the only way of measuring it. So we proposed, like I mentioned, that the public might possess this kind of deep-seated, gut-level trust in government and that this implicit trust might run contrary to what people are openly saying when they’re asked an opinion surveys about whether they trust government.

Grossmann: They believe that children learn to implicitly trust government.

Nicholson: Back to the childhood socialization literature, it’s about the overall political system and orientations toward it and those very kind of early idealized depictions of the political system. So we thought it would matter in that way.

Grossmann: Their new measure uses games like those to measure implicit racism.

Nicholson: We did essentially rapid categorization, sorting task and the idea is that you’re given a referent, which is government, and then you’re given different attributes that could be relevant to government and one was trust and one was distrust.

So, we had a list of words that were synonymous with trust and list of words that were synonymous with distrust and in the categorization task, they were at one point asked to pair the trustful words with government and then in another round they were asked to pair the distrustful words with government.

And we randomly assigned whether they got one or the other task first, we found that that really did make a difference and we found that most people were more likely to quickly associate trustful words with government than they were distrustful words. And because this task happened so quickly and these responses are spontaneous, it’s like playing a video game where you’re not giving it much thought, you’re just responding in the minute or, I should say, in the split second.

So the ideas that we’re capturing are things that people aren’t even really necessarily able to articulate.

Grossmann: They found that the two types of trust were unrelated and the old measure still matter for lots of attitudes like government spending, but they found implicit trust mattered more for system support.

Nicholson: We also in the study included their measures of explicit trust, the trust that people provide in surveys, and we found that it was relevant to all of the different types of dependent variables we’re looking at. You know, it was relevant to system support, it was relevant to government spending, it was relevant to crisis support in government so it was a pretty consistent predictor across the board. Where we did find differences in terms of whether the implicit or explicit measure mattered was in this area of government spending and we had come to that thinking that it would likely not, the implicit measure, likely would not have an effect there.

Grossmann: They also found implicit trust was more important when trust was asked about in the context of a crisis.

Nicholson: Question, more or less, just asked them to imagine crisis situation and this is, obviously, a limitation because it’s nothing like being in the real thing. Nevertheless, we found that the explicit responses even exhibited higher support. So when we just look at the responses, they’re higher than kind of the normal trust in government question. So we do think that the question did pull from people a little bit this kind of crisis mindset and we saw, like I said, more a trustful response to those items than we did your standard trust in government items. And then we found our implicit measures mentioned and it was a significant predictor of trust or explicit trust in government during crisis events.

Grossmann: Nicholson says the new measure is also independent of partisanship.

Nicholson: Contrast to explicit trust which can vary quite a bit depending on which party controls the White House. When the Republican Party is controlling the White House, Democrats are less trustful of government and when the Democrats are in control of the White House, Republicans are less trustful. So explicit trust in government very much does vary depending on which party is in power. Plus it did not vary by partisanship. We looked at differences between Democrats and Republicans on implicit trust, there were no significant differences. We looked at differences between liberals and conservatives, there was no significant differences.

Grossmann: The findings left him a bit optimistic.

Nicholson: I think our findings would be comfort to policymakers and political officials insofar as low trust in government does not necessarily mean that support for the political system is eroding. Again, it means that the specific reference, the people who are holding political office or types of government programs, they may be held in low disregard…

Grossmann: But Webster’s not holding his breath for improvement. He sees incentives for politicians to increase anger and distrust.

Webster: The way that politics is going, increasingly viewed as some game in terms of my win is your loss and vice versa. And people are increasingly trying to provoke anger because in terms of electoral strategy, we know that an angry voter is a loyal voter. So there are incentives to create and stoke anger and as far as what shows here, this has this broader sort of institutional result here. And so I’m not optimistic about trusting government increasing. I think if the trends that we see now continue, I think trust in government is likely to get even lower. And so I think future research should go and try to look at the ways we can reverse the unfortunate consequences of this rising anger. But, unfortunately, I’m quite pessimistic about it.

Grossmann: He says the next two elections are likely to bring even lower trust.

Webster: I think 2018 will be a very angry election. And I think 2020 may be even worse. 2020 will be an election that you can tune out as much as you can. I think you’d probably be doing pretty well for yourself; that’s gonna be pretty nasty.

Grossmann: And Nicholson agrees that we may see further declines in explicit trust.

Nicholson: Trust in government is just declining more and more and aside from the strongest supporters of the president, which might have slightly higher trust in government than its detractors, that doesn’t look to be improving any right now. Again, if there was some sort of crisis in the United States we could see it, temporary, return to higher trust levels, but none would wish that, of course.

Grossmann: So where do we go from here? Webster is looking at which information sources matter most for distrust.

Webster: Distrust goes down more when we’re made angry a co-partisan or an opposing partisan, what is the role of the media in doing this? So I think those are really promising areas to look for in future research. And then of course the big normal question is, to the extent that this declining trust in government is bad for democratic performance, is there anything we can do to reverse this decline in trust? I think sort of finding ways to sort of mitigate the unfortunate effects of anger to something that would be very well served to study in the future.

Grossmann: Nicholson wants to compare distrust in the U.S. and China.

Nicholson: We’re going to be planning a survey, I think, to examine trust in government and China, see how that varies or how that differs from a democracy like the United States. And there’s other questions that could be asked in looking at the authoritarian regime compared to democracy, of course. In the United States, I would love to see other people do research on implicit trust in government and replicate our study.

Grossmann: And Webster wants to find out the relationships between our broad personality traits and our momentary emotions.

Webster: The things that I find most interesting is looking at the intersection between personality and emotion. I was sort of alluding to this earlier, but if personality is some sort of stable dimension of how people just behave, like, I am an angry person or I’m a relatively calm person, but that doesn’t mean I can’t experience temporary deviation from that state.

Even very calm people are made angry sometimes, imagine somebody runs into the back of your car when you’re out driving. That’s gonna make you angry. So I think looking at how anger interacts with personality, the real interesting way to push forward both the literature on emotion and politics and personality and politics.

Grossmann: There’s lot more to learn. Political Research Digest is available biweekly from Niskanen Center and on iTunes. I’m your host, Matt Grossmann.

Thanks to Steven Webster and Stephen Nicholson for joining me. Join us next time to find out whether Americans are becoming tribal with identity politics trumping all.

Photo Credit: Ted Eytan under CC by SA 2.0.