two kinds of evidence suggest that high-income Americans do not hold stronger views of policy issues or consider such issues to be more important to them. First, surveys occasionally ask respondents to indicate their policy preferences and then to report how important that policy issue is to them. The 2004 American National Election Study, for example, asked a series of questions about gun control, government health insurance, defense spending, aid to black people, and environmental protection. Each policy item was followed by a question asking, “How important is this issue to you personally?” with five options ranging from “extremely important” to “not at all important.” The patterns differed slightly across these five issues, with gun control eliciting slightly greater “personal importance” ratings from low-income Americans and health insurance slightly greater importance from those with high incomes. But averaged across the issues, low-, middle-, and high-income respondents expressed nearly identical levels of importance; the percentage indicating these issues were “extremely” or “very” important to them was 58, 61, and 58 percent for low-, middle-, and high-income Americans, respectively. The second indication that high-income Americans are no more fervent in their policy preferences than those of more modest means comes from the subset of my data that contains measures of both direction and strength of preference. One hundred sixty of my survey questions ask respondents to indicate not only their support or opposition to the proposed policy change, but whether they support or oppose that change “strongly” or only “somewhat.” The bottom panel in figure 3.7 shows no difference across income levels in the propensity of respondents to say they “strongly” as opposed to “somewhat” favor or oppose a given policy.

Martin Gilens, Affluence and influence: economic inequality and political power in America, Princeton, NJ Oxford, 2012, p. 110