Download data and study materials from OSF
Principal investigators:
Suzanne Mettler
Cornell University
Email: E-mail: sbm24@cornell.edu
Homepage: http://government.arts.cornell.edu/faculty/mettler/
Matt Guardino
Providence College
Email: E-mail:mguardin@providence.edu
Homepage: http://www.providence.edu/polisci/faculty-profiles/Pages/mguardin.aspx
Sample size: 526
Field period: 02/21/2008-02/27/2008
Much of contemporary U.S. social provision constitutes a "hidden welfare state," a cluster of policies that uses the tax code to channel benefits disproportionately to affluent citizens. While politically and socioeconomically consequential, these policies are complex, arcane and opaque, and have not been subjects of high-profile partisan debate and news coverage. Employing an experimental design, we find that information about the mechanics and distributional consequences of tax expenditures promotes opinion expression, especially among those with at least moderate levels of general political knowledge. Overall, once citizens are informed of the distributive effects of these obscure policies, they become less supportive of those that benefit especially the affluent. Moreover, policy-specific information appears to help citizens align their preferences with their predispositions, including values and material interests. We suggest that if political elites and news media offered more information, public opinion toward the hidden welfare state would be more firmly grounded, and considerably less favorable.
1) Providing citizens with specific facts about the shape and implications of tax expenditures will spur policy preference articulation among those who would otherwise be at a loss to express opinions.
2) Opinion formation will be moderated by general political knowledge. In other words, a baseline of general knowledge will facilitate individuals' comprehension and incorporation of specific policy facts. Thus, subjects with higher levels of general knowledge will be more likely to respond to specific information about tax expenditures by articulating opinions than will those with lower levels.
3) Policy-specific information will facilitate attitudinal shifts in tax expenditure support that cohere with respondents' material interests, and/or with their ideological and political values. Specifically, low- and middle-income subjects, those who identify as Democrats and/or liberals, and those who express greater concern about rising economic inequality will report lower levels of support for programs that primarily benefit the affluent, and higher levels of support for a program that primarily benefits low-income citizens. Conversely, high-income subjects, those who identify as Republicans and/or conservatives, and those who express little concern about rising economic inequality will exhibit the reverse pattern of attitude change when exposed to specific information about tax expenditures.
Level of information provided for each of three tax expenditure policies (Home Mortgage Interest Deduction, Retirement Savings Contribution Tax Credit, Earned Income Tax Credit):
Group 1: First Stage --- Policy name. Second Stage --- Policy name + basic description of program.
Group 2: First Stage --- Policy name. Second Stage --- Policy name + basic description + distributive impact (i.e. percentage of benefits flowing to various income groups) in textual and graphical form.
Group 3: First Stage --- Policy name + basic description. Second Stage --- Policy name + distributive impact in textual and graphical form.
1) Provision of both descriptive and distributive information increased the propensity to express opinions about all three tax expenditure policies among subjects generally.
2) Subjects with moderate (as opposed to low or high) levels of general political knowledge showed the greatest propensity to articulate opinions after being exposed to specific information about the upwardly distributive policies (the Home Mortgage Interest Deduction and Retirement Savings Contribution Tax Credit). Subjects with high levels of general political knowledge showed the greatest propensity to express opinions after being exposed to specific information about the downwardly distributive policy (the Earned Income Tax Credit).
3) Provision of distributive information about the upwardly distributive policies caused subjects generally to become less supportive of and more opposed to these policies. Provision of distributive information about the downwardly distributive policy caused subjects generally to become more supportive of and less opposed to this policy.
4) The patterns in Finding #3 about the upwardly distributive policies were strongest and most consistent among low- and middle-income subjects, self-identified Democrats and liberals, and subjects who expressed greater concern about rising economic inequality.