Liberal Bias

A group of conservative psychologists headed by John Chambers has been fighting back against what they perceive as a liberal bias among social psychologists. Their latest paper demonstrates some systematic errors in assessing social mobility in the United States, with liberals making greater errors than conservatives. The authors refer to these errors as “distorted along ideological lines.”

Social mobility refers to the upward or downward movement of individuals between social classes. It is usually measured by comparing different generations within the same family, called intergenerational mobility. If the United States were a land of great economic opportunity, there would be only a modest relationship between people's wealth and that of their parents. However, studies have shown that there is less social mobility in this country than in most other industrial democracies.

Until recently, most scholars believed that social mobility in this country had declined in the last few decades. This was an inference based on the well-established fact that income inequality has increased dramatically since 1980. Since it is known that countries with the greatest inequality have the lowest social mobility, it was presumed that social mobility had been declining as well. However, in 2014, a research group led by Raj Chetty published an extensive study of the intergenerational mobility of Americans born between 1971 and 1993 using income tax data. They found that social mobility has been relatively constant over the time period covered by the study. The authors note that much of the increase in inequality has occurred at the extreme upper end (“the 1%”), which doesn't have much effect on social mobility. 

Chambers and his colleagues did an internet survey of 410 American adults in which they were asked to estimate what percentage of Americans born in the early 1980s whose parents were in the top, middle and lower thirds of the income distribution had migrated to each of the other two thirds of the distribution by the time they were in their late twenties. The accuracy of their estimates was measured against the Chetty, et al., data. They were also asked to indicate, on a 5-point scale, whether they thought social mobility had increased, decreased or stayed the same over the past 40 years. Finally, they were asked to state their political idelogy on a 5-point scale running from “strongly liberal” to “strongly conservative.” 

Here are the main findings (which were replicated in a followup study):

• Most respondents thought there was less social mobility than had actually occurred. Although all three groups underestimated social mobility, liberals underestimated it more than moderates and conservatives did. With regard to people whose parents were in the middle class, liberals underestimated the percentage that were upwardly mobile, while conservatives underestimated the percentage that were downwardly mobile. These findings are illustrated in this chart.
  
  

• The majority (56%) thought that social mobility had declined over the past 40 years, with only 15% answering correctly that it was stable. While the majority in all three groups thought mobility had declined, a higher percentage of liberals gave this incorrect response than moderates or conservatives.

The behavior of the liberals in this study exhibits motivated reasoning. They tend to believe what their ideology tells them, even though in this case it isn't true. Their errors were consistent with the liberal critique of this country that it does not permit enough social mobility. (Conservatives were also guilty of motivated reasoning when the underestimated the percentage of middle class people who were downwardly mobile.)

Some social psychologists and journalists have stated or implied that politically motivated reasoning is more prevalent among conservatives than liberals—the assymetry thesis. Chambers and his colleagues rightly point out that their results contradict the assymetry thesis. It is likely that all people engage in motivated reasoning, the size and direction of their errors depending on the issue at hand. Since liberals favor progressive change while conservatives value the status quo, liberals may exaggerate the seriousness of current problems, such as limited social mobility. Chambers is free to call this “distortion” as long as he applies a similar label to the errors of conservatives.

(Note to researchers: If you want to find “distortion,” it's probably best to present participants with a cognitive task similar to the one used by Chambers—a task that is unfamiliar, and for which everyday life provides little help in arriving at a correct answer.)

Let me tell you what is not the intended take-home message of this post. Just because the public underestimates the amount of social mobility in this country, that does not mean that the current situation is satisfactory. The Horatio Alger myth is still a myth. Miles Corak demonstrated in 2006 that social mobility in the U.S. is substantially lower than in Canada and seven of the eight European countries in the study. Only the U.K. has a (slightly) more rigidly stratified society than the U.S. As one pundit suggested, if you want to live the American dream, move to Finland. Social mobility also varies dramatically by region with the U.S., with mobility being lowest in the southeastern states. (See this PBS interview of Raj Chetty.) Emmanuel Saez, one of the co-authors of the Chetty study, put it this way: “The level of opportunity is alarming, even though it's stable over time.”

Furthermore, the fact that inequality is increasing means that low social mobility has more serious consequences. As the rungs of the economic ladder move further apart, the economic penalty paid by those who are stuck in the middle and lower classes becomes greater. Quoting Saez again, “The consequences of the 'birth lottery'—the parents to whom a child is born—are larger today than in the past.”

You may also be interested in reading:

In Denial

Herding Cats

Is Democracy Possible?  Appendix