By Scott Burris
In recent posts, I have been pointing to research that suggests that government intervention for public health is actually rather popular as a general matter. Now comes a neat paper that takes on the question of whether politicians actually know what their constituents want. I read it as further evidence that our politics is being shaped by a lot of well-supported anti-government noise-making that has been allowed to flourish unchallenged.
The paper in brief: the authors surveyed candidates for state-level legislative office, and used a technique called multi-level regression and post-stratification (MRP) to localize opinion poll data to legislative districts. They then compared what candidates think their constituencies believe on key issues (health care reform, gay marriage, welfare reform) with what the polls say their constituents believe. They find that both conservatives and liberals significantly overestimate the conservatism of the people who elect them:
“In districts where supporters of these policies outnumber opponents by 2 to 1, liberal politicians appear to typically believe these policies enjoy only bare majority support while conservative politicians typically outright reject the notion that these policies command widespread support.”
The paper is worth reading for its findings (and to allow you to personally assess its limitations – this has not yet even been peer reviewed.) A more detailed summary with some of the charts is on Dylan Matthew’s Washington Post blog.
In their discussion, the authors suggest there is a “folk theory” among conservatives that the public is considerably more conservative than it appears, a theory they see in campaigns ranging from Nixon’s “silent majority” to Sarah Palin’s “real America.”
“Liberal politicians exhibit a similar conservative perceptual bias that, although less severe, contributes to the overall pattern that politicians typically overestimate opposition to the policies we study by over 10 percentage points on the policies we examined. Perversely, these differences mean that liberal politicians were broadly congruent with their constituents but in many cases unaware of this congruence, while conservative politicians were quite often incongruent but nearly never cognizant of this fact.”
Here’s a particularly telling example, on the question of “[a]bolish[ing] all federal welfare programs.”
“Very few Americans agree with this extreme statement – a complete dismantling of the entire American welfare state – yet American politicians believe that support for this sentiment is extremely widespread. Indeed, even liberal politicians typically believe that support is well above 25% in their districts, while conservative politicians typically perceive opinion on this issue closer to 40%. The true national mean is 13%, considerably more liberal than politicians of all stripes believe.”
The relationships among actually felt attitudes and preferences, polls, politician’s perceptions of preferences, and the media’s characterization and dissemination of “public opinion” are all complicated by how we all think, get our information and process it. But my perception is that all around me, claims that Americans don’t like big government or regulation or active health promotion are too often allowed to pass as true (not least by the supposedly “liberal” media), and the politicians who should be standing up in support for action are caviling and cowering in the face of perceived high levels of public opposition.
Those who make those claims repeat them endlessly, doggedly, effectively, and our failure to challenge them allows them to be perceived as accurate. The prescription for those who favor collective action for national well-being is to start making more noise. It’s true that even high levels of popular consensus may not matter in our money-driven system (background checks, anyone), but fake consensus is an actionable target that still matters: Our silence allows the perception of a false anti-action consensus; our refusal to be silent any more can help shift perceptions.