Trading Off Openness: Information, Incentives, and Preferences for Immigration Policy

We present three future policy scenarios characterised by different levels of stringency and openness of immigration to individuals in three countries: Italy, Germany, Hungary. For each of the three scenarios, we ask respondents to indicate whether they would support the proposed migration policy or prefer to keep the current one, which we label as the status quo. In our experiment, we manipulated the amount of information provided to the respondents. The control group only visualizes information on the immigration policies. The treatment groups are also informed on the demographic and economic consequences of the status quo and the alternative immigration policies. In Treatment 1, we present the implications in terms of working-age population and GDP per capita in 2050; in Treatment 2 in terms of
personal income.
After respondents are exposed to three sets of binary choices between maintaining the status quo and opting for the alternative scenarios and express their preference, we also collect incentivized decisions to gauge revealed preferences for supporting immigration. We identified an organization that is involved in both international cooperation programs in countries of the global South and in migrants' integration programs in destination countries.

2025-04-27

2025-07-31

The primary outcome variables include stated policy preferences in three different migration scenarios caraterised by different levels of stringency/openess to migration and an real-stake decisions in support to immigration.

We measure self-reported preferences computing a dummy variable equal to 1 if respondents prefer the alternative policies compared to the status quo. We inform respondents that migration will caracterise the following twenty-five years (until 2050). The status quo maintains the number of working-age migrants allowed to enter every year at the current level. The three alternative future policy scenarios are the following: Closing borders, whereby the annual inflow of working-age migrants allowed to enter will be set to zero. Lower Immigration, whereby the inflow of working-age migrants who are allowed to enter every year will be lower than the current level. Higher Immigration, whereby the inflow of working-age migrants who are allowed to enter every year will be higher than the current level. in each of the three scenarios, respondents should indicate if they prefer the alternative policy or the status quo.

We measure incentivesed preference using information on donation to an organization that works in support to the integration of immigrants and their participation in the labour markets of host countries. We give survey respondents a 1 euro endowment and offer them the opportunity to donate part of it to support the two different programs. One program aims to provide support to residents in developing countries, through initiatives that improve their living conditions and increase employment opportunities in their country of origin. The other mission aims to increase the integration of immigrants and their participation in the labour markets of host countries.

We will analyse the effect of Treatment 1 and Treatment 2 on both outcomes. Additionally, we will examine heterogeneity in treatment effects by political orientation, beliefs about zero-sum thinking, and attachment to socio-demographic and cultural identity. We measure beliefs about zero-sum thinking through the answers to the question "Please indicate your level of agreement with the statement, using a scale from 0 (Strongly disagree) to 10 (Strongly agree): If someone succeeds in life, it often happens at the expense of someone else". We measure beliefs about zero-sum thinking through the answers to the question "Please indicate your level of agreement with the statement, using a scale from 0 (Strongly disagree) to 10 (Strongly agree): It is important to preserve the socio-demographic and cultural identity of the country" .

The secondary outcome variables include post-treatment beliefs on the impacts of immigration, to understand if respondents trust our treatments and update their beliefs accordingly; general and self-reported questions on attitudes towards migrants; responses to the open question on possible solutions for the ageing of the population in their own country; beliefs about the share of migrants in their own country. This outcome helps us determine whether
our treatments—by highlighting the impending decline of the native population—backfire by exacerbating concerns about the shrinking ratio of natives to foreigners.

The experimental design includes individual-level randomization. We randomly assign individuals to three groups. The control group only visualizes information on the immigration policies. The other two groups are also informed on the demographic and economic consequences of the status quo and the alternative immigration policies. One of these two groups visualzed the economic consequences in aggregated form,
namely as per-capita GDP. The other group visualize them in a personalised way, in terms of own income level. We retrieve own income to all respondents through a pre-treatment question.

Randomization done by Qualtrics.

We will have three samples: one in Italy, one in Germany and one in Hungary. Within the sample, the randomizations are conducted at the individual level.

No

3000 participants in Italy; 3000 in Germany; 3000 in Hungary

9000 adult participants

3000 participants by treatment arm