Perceptions on Intergenerational Mobility and Preferences for Redistribution in Spain

We conduct three waves of online surveys, Wave A, the control group, and Waves B and C, the treatment groups.

Two animations on low intergenerational mobility will be shown to the respondents of Waves B and C. Respondents of Wave A do not receive the animated information.

2024-02-19

2024-03-30

1. Perceptions on personal and household income positions in the country and on intergenerational upward mobility.
2. Policy preferences.

1. Perceptions
1-1. Perceptions on personal and household income positions in the country
We ask a question on their perceived relative positions of personal and household income in the national income distribution. We also ask questions on last year’s personal and household income, of which the categories are based on actual income decile distribution in the country (EU Statistics on Income and Living Conditions, 2022). Combining those questions, we show stylised facts on Spanish respondents’ misperceptions on their own income positions in society.

1-2. Perceptions on upward mobility
To elicit respondents’ beliefs on intergenerational upward mobility, we introduce an image of stairs to the question, in the same manner as preceding research. Hypothesising 500 families that represent the country's population, we ask the perceived probability of upward mobility of the children coming from poorest families. Specifically, using an image of stairs, we ask respondents how many children coming from the poorest 100 families will belong to different income groups, once they grow up (respondents need to fill out empty fields to indicate their views). We then redo the same question of upward mobility conditional on children’s effort and talent.

2. Policy preferences
The main questions on policy preferences are as below.

2-1. Government budget allocation
We ask preferred budget allocation on ⅰ) Defense and National Security, ⅱ) Infrastructure, ⅲ) Education (early childhood, primary and secondary education), ⅳ) Education (tertiary education), ⅴ) Old age and disability pensions, ⅵ) Social Insurance and Public Assistance, and ⅶ) Health. The respondents have to divide the budget in to 7 categories and the total percentile of the budget must sum to 100.

2-2. Income taxes
To elicit income tax preferences, we ask respondents to move taxation sliders, which correspond to preferred income tax rates on each income group, in a way that they view fair. Income groups are the top 1%, the next 9%, the 40%, and the bottom 50%. Notice that respondents are constrained to current level of revenue in their country, and to build up this constraint, actual statistics, such as GNI and income tax revenue statistics, are utilised behind the question.

In addition to the question above, we ask policy preferences questions, such as support for the Estate Tax, equality of opportunity policies. For the details of the main questions, please see the Supporting Documents and Materials, which is available upon request (by clicking the button).

We conduct an online survey to collect information on perceptions on intergenerational mobility and on relative income positions in the country, and preferences for redistribution amongst a representative sample of the Spanish population. We attempt to clarify correlations and causations between upward mobility perceptions and preferences for redistribution policies. The survey design builds on Alesina et al. (2018), which shows treatment group respondents two animations on low intergenerational mobility.

Based upon the aforesaid information-provision design, respondents are randomly split between one control group and two treatment groups, Waves A, B and C respectively. Respondents in treatment groups, Wave B and C, are shown two animations on low intergenerational mobility, those in the control group, Wave A, are shown no information. We also run a follow-up survey to the treated, Waves B and C, in order to check persistency of treatment effects. Respondents are recontacted approximately 7 days after they have completed the main survey.

For the details of randomisation strategy and key questions for our study, please see the Supporting Documents and Materials, which is available upon request (by clicking the button).

Planned date for data collection.
■Trial Start Date
Wave A (Control G): 2024-02-13
■Intervention Start Date
Wave B (Treatment G): 2024-02-19
Wave C (Treatment G): 2024-03-18
■Intervention End Date
Wave B (Treatment G): 2024-02-29
Wave C (Treatment G): 2024-03-28
*Follow-up survey will be conducted for Waves B and C, approximately 7 days after they have completed the main survey.

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

We randomize participants (individuals) in control and treatment groups.

No

One country

2,000 respondents

500 respondents for control, 1,500 respondents for treatment.