The Partisan Trust Gap

The following is a guest post by Mike Sances:


The causes of low trust in government have been a recurring topic of discussion on The Monkey Cage. Which explanation is right is key to any prescriptions for increasing confidence in our political institutions. If it’s about a policy disconnect, perhaps major reforms will be needed to decrease the influence of interest groups. It If it’s about deeply held feelings of alienation, then maybe what is needed is a massive educational campaign about the virtues of American government. If it’s the economy, then maybe we can just sit back and wait for the economy to grow again.

The concern about low trust in government is not new, and neither is this debate. In a classic exchange in a 1974 issue of the American Political Science Review, Arthur Miller and Jack Citrin debated why trust in government had declined so sharply in the late 60s and early 70s, and what would make Americans trust government again. Miller argued that the decline was about policy dissatisfaction: Americans were turned off by centrist policies, and wanted something less blasé. Only liberal, activist government moving aggressively to tackle big social problems would reverse the decline in trust.

Citrin was unconvinced, and countered that cynicism was not monolithic. Instead, Citrin argued there were three types of political cynics: “ritualistic” cynics, who just give cynical survey responses to conform to social norms of skepticism; “partisan” cynics, who just don’t like the current officeholders; and “alienated” cynics, who are genuinely turned off from government. Citrin couldn’t identify the proportions of each type exactly. But by showing that Democrats were more trusting of government under Johnson, and that Republicans were more trusting of government once Nixon came into office, Citrin demonstrated that there was indeed a nontrivial amount of partisan cynics in the electorate. But he concluded that what would ultimately raise trust was a booming economy: “results” in his words, rather “than the adoption of some particular program or ideological orientation.” As John Sides has shown, Citrin’s prediction had a good deal of truth to it. However, it turns out his insight about “partisan cynics” was just as prescient.

Since the trust in government question was first asked in the late 50s, it has always been the case that members of the president’s party say they are more trusting of the federal government—an effect that is only getting bigger over time. Below, I plot the proportion of respondents saying they trust the government “always” or “most of the time,” over time. I use the same Pew data from John’s earlier posts, but I here disaggregate responses by party identification (as, in fact, anyone can do at that Pew web page). D’s and R’s indicate proportions from individual polls in the Pew dataset, and the vertical dotted lines indicate a change in the party occupying the White House.

There has always been a “trust gap” between members of the president’s party and the opposition. While members of both parties became less trusting from the 60s and 70s, Republicans’ trust was strikingly high under presidents Reagan and Bush II. And while Republicans’ trust also fell under Nixon, it did so at a slower rate than among Democrats. This same pattern of “catching up” by Republicans can be seen in Bush II’s second term.

What also stands out is that the gap between partisans is getting bigger. In the figure below, I take the absolute difference between the proportion of each group of partisans who say they are trusting always or most of the time, and plot it across time. The points are individual polls, and the red line is a lowess fit.

This figure confirms the earlier figure: there is always a trust gap, and it is getting bigger. In 1958 it was about 8 points. In the 80s, it got as high as 22. Under Bush II, it peaked at 35 points. And under Obama, it’s still considerably high, in the low 20s in the most recent data.

So, what will make people trust government again? The numbers suggest that, in addition to a booming economy, having some political power also helps. Why this is the case, and increasingly so, is surely an interesting question. But it’s one we won’t be able to answer until we stop pining for the “good old days,” when trust was universally high.

7 Responses to The Partisan Trust Gap

  1. Rob April 14, 2011 at 2:18 pm #

    Though it’s not the heart of Mr. Sances’s essay, his comments in the introductory paragraph about prescriptions for increasing confidence in our political institutions are revealing. For example, he states, “It [sic] it’s about deeply held feelings of alienation, then maybe what is needed is a massive educational campaign about the virtues of American government.”

    By analogy, if people have low trust in used car dealers, what’s needed is a massive educational campaign about the virtues of used car dealerships.

    You want people to have more trust in government? Maybe the government should try being more trustworthy. When a President repeatedly tells people that under his health care bill, “if you like your health care plan, you can keep your health care plan,” and that turns out to be a massive oversimplification and proves for many people to be incorrect, that undermines confidence. When a President says, as President Obama did yesterday, that if Medicare costs rise faster than a set rate, the Independent Payment Advisory Board will cut Medicare expenditures to the extent necessary “by further improving Medicare,” is it any wonder that people who intuit that “improving” is slick salesman’s lingo for “diminishing” will lack confidence in government?

    In “Jaws,” after observing the enormity of the shark, Roy Scheider says, “You’re gonna need a bigger boat.” Well, however massive an “educational campaign” you think we may need about the virtues of American government, you’re gonna need a bigger campaign.

  2. Andrew Gelman April 14, 2011 at 3:10 pm #


    1. Nice graphs!

    2. You write of the question about trust: “it’s one we won’t be able to answer until we stop pining for the ‘good old days,’ when trust was universally high.”

    I don’t understand what you’re saying here. Why can’t we simultaneously (a) pine for the good old days (according to your graph, trust was in the 70-80% range during the Eisenhower-Kennedy years), and (b) try to study the decline. I don’t see why not-pining is a requirement for going forward.

  3. John Sides April 14, 2011 at 3:23 pm #

    Rob: I think I introduced that typo when I was formatting Mike’s post for publication. But more importantly — and although Mike can certainly speak for himself — I don’t think you’re really arguing with anyone here. Mike is not arguing that we need a massive educational campaign. He is simply pointing out that the origins of people’s attitudes toward government have implications for the phenomena that would affect or change their attitudes. You seem to be reading in some sort of normative pro-government position in the post, but none exists.

    Andy: I didn’t write the post, but maybe Mike will weigh in.

  4. Mike Sances April 15, 2011 at 12:56 pm #

    Rob: I wasn’t suggesting that an educational campaign is needed or desirable, just that some people might think so once they decide that deeply rooted alienation is a cause of distrust. Your basic point about politicians lying is an interesting one, though. Unfortunately, I don’t think the amount of lying by politicians has varied that much, so lying can’t explain these graphs. Right?

    Andrew: I phrased that poorly. What I was trying to suggest is that there’s been too much focus on the dip from the late 60s, and that because of this focus, there are more systematic patterns that haven’t been looked at yet.

  5. Rob April 15, 2011 at 3:26 pm #

    Mike, it’s probably true that lying by politicians hasn’t varied that much, though that’s something that’s probably impossible to measure empirically, and it’s also the case that as politicians’ communications have become more sophisticated, the tendency to lie by implication and oversimplification may have increased the total quantum of what could be regarded as lying.

    But aside from simply considering the quantum of lying, we should factor in the possibility that people may have grown more skeptical over time and less willing to assume the trustworthiness of government. Watergate may have been a turning point in this evolution. Similarly, people are probably less willing to accept at face value advertising claims and assertions by businesses and perhaps even organized religions. We live in an increasingly skeptical age. As persons trained in a liberal arts education to judge everything we read and hear skeptically and to constantly ask ourselves, “Is this really true? Has the author supported his or her statement?,” we should welcome this critical faculty in the general population. Right? And that would explain a decline in confidence in government even if the quantum of lying remained constant.

    If people are decreasingly homophobic and racist over time, we regard that as progress and don’t generally imagine the decline in those attitudes might warrant a massive educational campaign to reverse the trend. If people have less confidence in faith healers than they once did, ditto.

    John Sides is right. I did perceive a normative pro-government position in your post, and pace John I still think that’s a fair reading of it. You may not have intended to make that point, and whether you did or not it’s a perfectly respectable view to hold, but I believe it’s useful for each of us to be introspective about what unconscious biases we may be bringing to our thinking and writing–and that’s especially true for those who aspire to be social scientists and wish to perform and present their analyses dispassionately. I wish you good luck and continued success in your academic career.

  6. jorod April 19, 2011 at 9:50 pm #

    I wonder if there is any correlation between this phenomena and failure to represent the taxpayers?


  1. “Cognitive Madisonians” and Congressional Approval | Rule22 - May 12, 2011

    […] is a sentiment many academics echo (in my experience).  For example, Mike Sances (guest posting over at the Monkey Cage) recently asked “if [declining trust] about deeply held feelings of alienation, then maybe what […]