Immigration and Redistribution

We run online surveys in six countries. We recruit participants through professional survey companies to ensure that our sample is representative of each country's population along key dimensions - gender, income, and age. Only natives (non-immigrants) between 18 and 70 years of age are allowed to take the survey.

2017-10-11

2018-02-28

Misperceptions of immigrants characteristics, views on immigration policies, views on redistribution policies, real outcome (donation to charity).

Misperceptions are defined as the difference between perceived and actual statistics about immigrants (share of foreign-born, unemployment, poverty rate etc.).
Views on immigration policies are captured by answers to the following questions: how much immigration there should be, whether the government should care equally about immigrants and natives, when immigrants should be eligible for benefits, when they should be able to get citizenship and vote, and when they would be considered to be truly part of the country.
Views on redistribution policies are captured by answers to the following questions: preferred progressivity of the tax system, preferred allocation of budgetary funds, should the government increase funding for schooling, housing, and income support programs for low-income people, whether income inequality a problem, should the government care about income inequality.
Finally, we tell respondents that they have been automatically enrolled in a lottery to win $1000. Before they know whether they have won or not, they need to commit to donating none of it, part of it, or all of it to one or two charities (targeted towards low-income adults or children in general and not concerned with immigrants particularly and popular and well-perceived in each country). We use the amount respondents wish to donate as (real) measure of support for private redistribution.

In the surveys, we collect respondents’ detailed, quantitative perceptions about immigrants’ numbers, origins, religion, net benefits received, unemployment rate, poverty, and education. We also ask detailed questions about support for various tax and transfer policies, social insurance, and migration policies. To capture respondents' perceptions we use various visualization tools, such as sliders and interactive pie charts. In addition, we perform four randomized experimental treatments. The first treatment consists in randomizing the order of questions about immigration and about redistribution: half of the respondents are shown the immigration block before the redistribution block, the other half answer questions in the opposite order. In the remaining three treatments we show respondents information about the true i) number, ii) origin and iii) “hard work” of immigrants in their country, respectively. Respondents are randomly split between one control group and three treatment groups. Respondents in each treatment group are shown only one piece of information, those in the control group are shown no information. In the U.S. we also run a follow-up to check persistency of treatment effects. Respondents are recontacted 7-10 days after they have completed the main survey.

Respondents are randomly assigned to control and treatment groups by the Qualtrics software.

We randomize participants in control and treatment groups within each country

No

6 countries

22,000 respondents

1,000 respondents for each group in each country