Abstract
- Do perceptions of intergenerational mobility in a country affect political opinions and preferences on wealth redistribution? The study shows that there is a correlation, but not causality.
In 1966, American author John Steinbeck (Nobel Prize winner in literature) conjectured that American society did not particularly support wealth redistribution policies because even the poorest people saw themselves as « temporary millionaires in misery. » Some 50 years later, Alberto Alesina, Stefanie Stantcheva, and Edoardo Teso tested this intuition in a more scientific way: is there really a link between support for redistribution policies and the perception of the ability to climb the income ladder in a society?
An overview
To answer this question, the authors began by conducting a large-scale survey in five countries: France, England, Sweden, Italy, and the United States. These five countries are interesting in that they have different cultures and institutions, and therefore different social perceptions. Nearly 10,000 people were surveyed. The survey assessed each respondent’s support for redistribution policies through questions about support for inheritance taxes and policies aimed at increasing opportunities for children from poor backgrounds. Each respondent’s perceptions of social mobility are assessed through questions about the chances of children from poor or modest backgrounds reaching the top of the ladder, as described in the example below.
The authors then use advanced statistical methods (randomized controlled trials) to assess the potential causal link between the variables in question. It should be noted that the study focuses on a specific definition of social mobility: the ability to move up the income ladder. It does not therefore refer to social mobility in terms of education, access to the best schools in the country, or access to political or highly intellectual professions. We will define this term below as « income social mobility. »
Some interesting initial findings
The study first quantifies differences in perception and how they relate to reality based on available data. A major advantage of this study is that it is standardized (the same questionnaire for all countries) and representative. This yields some interesting quantifiable findings. For example, Americans are « too optimistic » about income-based social mobility, while Europeans are « too pessimistic. » In other words, the Europeans surveyed give figures on social income mobility that are lower than reality, while Americans do the opposite. The differences are statistically significant, even if, quantitatively, they are not very large. The graph below shows, for example, that the average American believes that the probability of a child inthe first quintile of the income distribution reaching the last quintile is 12%, when the actual probability based on available data is 8%. The average French person makes almost the opposite prediction.
Perceived probability (y-axis) versus actual probability (x-axis)
One might think that the question is too technical for the figures to be truly representative. However, since each respondent is asked the same question, it is interesting to note that the average American overestimates the probability, while the average French person underestimates it.
The authors highlight other interesting findings from this study. One well-known psychological bias, for example, is the fact that we as humans often tend to take excessive credit for our successes and blame external circumstances and luck too much for our failures. The study data allows us to test this hypothesis directly: by estimating simple relationships controlling for multiple individual factors, they find that the higher an individual’s income (or the higher their level of education), the more they tend to agree with the statements that « the main reason for being poor is a lack of effort » and « the main reason behind wealth is individual effort. »
Testing the causal link
As might be expected, the authors report a strong correlation between social perceptions of inequality and support for redistribution policies. But, as everyone knows, correlation does not imply causation. Political opinions could condition social perceptions, or a third variable could influence both variables.
In order to get an idea of a potential causal link, the researchers sought to use a « treatment » to affect only the « perception of social mobility » aspect, and then analyze whether the variation in perceptions due to this treatment could affect political opinions. The « treatment » consists of a short animation informing the respondent of « the results of recent research on the subject, « namely that « the chances of someone born into a poor family remaining poor as an adult are extremely high, « or that « the chances of a child born into a rich family remaining rich as an adult, like their parents, are extremely high. » It is administered randomly to only a portion of the sample of respondents. By statistically comparing the opinions of those who received the treatment with those who did not, it is possible to identify an effect on perceptions due to the treatment, as well as an effect on political preferences.
The results show that the treatment appears to change perceptions of social mobility and, for left-wing respondents, their preferences on redistribution policies. Right-wing respondents in the sample analyzed are less sensitive to this message and do not change their preferences, although their perceptions are altered.
Conclusion
In conclusion, while this study provides a solid quantification of the correlation between perceptions of social income mobility and preferences for redistribution policies, the results on the cause-and-effect relationship are more mixed. Only those on the « left » seem to validate the initial hypothesis. The added value of this study lies largely in the interesting statistics it provides on issues relating to social perceptions, thanks to this unique survey of nearly 10,000 people from five countries with different cultures. It provides relevant information for analyzing public support for policies aimed at reducing income inequality.
