News

Higher education is often viewed as an engine of social mobility. However, individuals from low-income families are often underrepresented in selective universities, which can lead to higher inequality in labor market outcomes after college. Affirmative action policies aim to reduce this inequality by increasing the representation of low-income and minority students in selective universities.

Previous research has found that affirmative action works at moderately-selective universities. Affirmative action at these institutions allows disadvantaged students to attend colleges with greater resources and higher graduation rates. This increases the likelihood that they earn a college degree, which improves their labor market prospects.

But things are different at the most selective universities. All applicants to these institutions are highly qualified, and those who are not admitted are likely to earn a college degree elsewhere. So, what role do affirmative action policies play at elite institutions?  

A new study by briq postdoc Germán Reyes, together with Cecilia Machado and Evan Riehl, sheds light on this topic by examining the impacts of a large-scale affirmative action policy at one of Brazil’s most selective and prestigious universities, Rio de Janeiro State University (UERJ). In the early 2000s, UERJ began reserving 45 percent of its admission slots for low-income and minority applicants, which dramatically increased the racial and socioeconomic diversity of its student body.

The researchers analyzed the academic and early-career trajectories of all UERJ students, focusing on the policy’s impact on two distinct groups. First, the “targeted” students who were directly admitted through the affirmative action program. Second, the “non-targeted” students who would have gained admission regardless of the policy.

They found that the affirmative action policy boosted the early-career earnings for the targeted group. Unlike the research mentioned above, the researchers did not find any impact on educational attainment. Instead, the earnings benefit was driven by access to a set of high-quality firms associated with the alumni network of the university. However, this income benefit faded as they progressed in their careers.

UERJ’s experience reveals that at elite universities, affirmative action helps disadvantaged students gain access to valuable networks that lead to high-wage initial jobs. But this benefit does not necessarily persist over time because disadvantaged students face additional barriers to career advancement in the labor market. While affirmative action and similar admission policies are crucial in addressing disparities in college access, they alone cannot overcome all challenges faced by underprivileged individuals.

Research has consistently shown that individuals who possess a realistic perception of their abilities tend to excel in life compared to others. The ability to accurately assess oneself, whether contemplating an apprenticeship or pursuing higher education, choosing between language or science disciplines, or determining the best-suited path, plays a vital role. The precision with which young people answer these critical questions is heavily influenced by their social environment. A comprehensive study, based on the briq family panel, which systematically tracks the development of children from nearly 600 families, confirms this notion.

The study highlights the challenges faced by children from disadvantaged backgrounds in accurately classifying their abilities. This poses a significant disadvantage when making pivotal life decisions, such as selecting a career path. On average, individuals with more precise self-assessment achieve greater success, enjoy higher incomes, and even experience better overall health.

Marble game unveils insights into self-assessment skills

To gauge the self-assessment abilities of primary school-aged children, the researchers devised an innovative game centered around marbles, where children had the opportunity to win toys. In this game, eight to nine-year-olds were tasked with rolling marbles into holes. After a training round, they were given the autonomy to determine the difficulty of the task by choosing the size of the hole. Opting for smaller holes led to greater rewards, but it also increased the difficulty and the likelihood of not winning. The results of the game demonstrated that children from higher socio-economic status families outperformed those from lower status families.

Nurturing accurate self-assessment through mentoring

To examine who this disadvantageous starting position can be rectified, the study authors randomly divided the children from educationally disadvantaged families into two groups: one received support through a mentoring program for a year, while the other did not. The program, known as “Balu und Du,” enlisted the assistance of young individuals, predominantly university students, who volunteered to sponsor a child. They spent several hours each week engaging in interactive activities, such as playing together, crafting, making music, or participating in sports. These activities enabled the children to make new experiences, learn about themselves and thereby improve their self-assessment.

After a year, the mentored group of students scored significantly better in the self-assessment game. The mentoring program even had such a substantial impact on their self-assessment that during this time these children had reached the same level as their peers from socio-economically advantaged families. This showcases the vital role of interactive support in fostering the development of a realistic self-assessment.

Long-term positive effects

The positive effects of the mentoring program have proven to be sustainable over the long term. Six years after the conclusion of the mentoring intervention, the researchers revisited the children, who were now teenagers, and asked them about their school performance. Comparing self-assessed performance with actual grades revealed that children with higher socio-economic status and those who had participated in the mentoring program consistently displayed significantly better self-assessment abilities.

Historically, the weakening of ties among extended family has been associated with development, urbanization and economic growth. The causal relationship underlying this link, however, has been unclear. Kinship ties may react flexibly to changes in incentives rather than being fundamental causes of economic outcomes.

A new study by briq postdoc Arkadev Ghosh, “Sam” Il Myoung Hwang and Munir Squires, using U.S. state bans on cousin marriage, supports the hypothesis that tightly knit family structures impede development. To measure cousin marriage, the authors utilize the excess frequency of same-surname marriages, a method derived from population genetics, which they apply to millions of marriage records spanning the 18th to the 20th century. Using census data, they first show that married cousins tend to reside in rural areas and hold lower-paying occupations.

The researchers then employ an event study analysis to investigate the effects of cousin marriage bans on outcomes for birth cohorts subjected to these regulations. They discover that such bans prompt individuals from families with high rates of cousin marriage to leave farms and migrate to urban areas, and they also gradually shift to higher-paying occupations.

Additionally, the researchers observe increased occupational and geographical dispersion —  individuals from these families reside in a broader range of locations and pursue more diverse occupations. The findings indicate that these changes primarily arise from the social and cultural implications of dispersed family ties, rather than genetic factors. Notably, the bans also lead to a higher number of individuals residing in institutional settings for the elderly, infirm, or destitute, suggesting reduced support from relatives.

The study (see pre-publication version here) has now been published in the Quarterly Journal of Economics.

In an interview with the German business weekly Wirtschaftswoche, Luca Henkel, PhD candidate at the University of Bonn and research assistant at briq, points out that the negative image of stock owners may be responsible for low stock ownership rates. Below is an English translation of the interview (see original German version here).

+++

You conducted a study to find out people’s opinion on stockholders. Probably not a good one, right?

We started from the observation that many people, especially in Germany, do not participate in the stock market, and we investigated the role played by opinions and prejudices about stockholders. In our study, we show that people perceive stockholders very negatively and that these views have a concrete influence on their decisions to invest in the stock market. Specifically, because most people think that stockholders are more selfish and greedy than non-stockholders, they do not invest in the stock market themselves.

In your study, you examined groups from the Netherlands and the United States. What can be deduced from this for Germany?

In Germany, we lack concrete data so far, but we are currently working on that. Surveys among our students at the University of Bonn, however, seem to confirm the findings from the Netherlands. So in our country, too, there is a very negative perception of stockholders.

That fits in with the notion of the German “Aktienmuffel” who shuns stocks in general. But you write that in the U.S., with 86 percent of respondents, even more people perceive stockholders negatively than in the Netherlands (81 percent). How can that be explained?

The two figures are not fully comparable due to limitations in the U.S. data. In the Netherlands, we surveyed a representative population group. In the U.S., the responses were collected via an online panel. This is not representative of the population as a whole, and significantly fewer people took part there – only around 400, compared to almost 3,300 in the Netherlands. This difference explains parts of the results, which are surprising at first glance.

Stockholders have a bad image either way. But stocks have nevertheless become more popular as an asset class over the past ten years, when interest rates were exceptionally low for a long time. Does that show up in your data?

No, older people have neither more positive nor more negative attitudes toward stockholders than younger people. But older people are more likely to let these attitudes guide their behavior. So young people who are skeptical about stockholders show a greater willingness than older people to invest in the stock market.

Does people’s own wealth influence how they feel about stockholders?

Half of the richest 20 percent in the Netherlands own no stocks at all. Around 40 percent, in turn, have no significant assets to invest in the stock market. However, one’s own wealth does not play a decisive role in the question of how one feels about stockholders. The perception is predominantly negative in all wealth groups.

Does the negative perception only apply to stockholders? Are bondholders or real estate investors more popular?

We have not investigated this. What we can say, however, is that ETF investors or those who invest in sustainable financial products have a better reputation. But they are still perceived more negatively than non-stockholders.

Some seem ashamed of being a stockholder. Your respondents fibbed to you about that, didn’t they?

That’s right. In our research in the Netherlands, we had the opportunity to match the information on stock ownership in our survey with their tax records, which we believe provide a more realistic picture than surveys. The result of the comparison was astonishing: One-third of the respondents who, according to their tax records, owned stocks, reported to us that they did not own any. Conversely, only two percent of respondents whose tax records did not show stock ownership stated that they were stockholders.

So because people think negatively about shareholders, they fib and invest less in the stock market. How can this mechanism be overcome?

Our results suggest that, above all, the opinion about stockholders is a major obstacle to individual investing. Neobrokers have recognized this. They address precisely these stereotypes with their advertising campaigns, in which they show everyday people on the couch or on the train trading stocks. The idea is to get away from the image of the stockholder seen in Wall Street movies. Lack of financial knowledge, which is often cited as the main reason for low stock ownership rates, is not the main problem, according to our data.

A large-scale multinational study shows that differences in universalism are a stronger predictor of variations in support for various policy views than traditional political economy variables such as income or equity-efficiency preferences.

Moral universalism is defined as the extent to which people exhibit the same level of altruism and trust towards strangers as towards in-group members. Universalism focuses on how a person’s altruism or trust changes based on the distance between them and the group they are interacting with, rather than one individual’s overall level of altruism or trust. Therefore, a universalist places less emphasis on group- or place-based identities than communitarians do.

The study by Benjamin Enke, Ricardo Rodríguez-Padilla, and briq Research Director Florian Zimmermann, published in the Review of Economic Studies, provides descriptive evidence that moral universalism predicts the prevailing ideological clusters across many rich Western countries. The analysis is based on internet surveys with a total of about 11,000 respondents from nationally representative samples in the United States, Australia, Germany, France, and Sweden.

The researchers developed a new method for measuring universalism. Participants are given a hypothetical $100 and are asked to divide it between two equally rich individuals: (i) a randomly selected member of a specific social (in-)group who lives in their own country of residence and (ii) a randomly selected person who lives in their own country of residence. Each participant is required to make multiple choices with varying types of (in-)group members. For example, in one question, a U.S. participant is asked to split a hypothetical $100 between a member of their extended family and a randomly selected person from the United States. The amount given to the random stranger across the different decisions makes up the universalism measure.

Structure of political ideology in Western countries

As a first step, the authors use their rich survey data to study patterns in political views. The data suggests that individuals’ policy views in affluent Western nations are highly correlated, despite notable variations in their electoral systems, political parties, and demographic makeup. People in one cluster generally desire government expenditure on foreign aid, affirmative action, environmental protection, welfare, and universal health care, while people in the other cluster support government spending on the military, police and law enforcement, and border control. But what determines who belongs to which cluster?

Universalism and policy views

The study documents that heterogeneity in moral universalism plays a central role in shaping the specific structure of policy views described above. Universalism is highly correlated with the respondents’ self-reported ideology on a left-vs.-right scale (see Figure B). Further, across all countries, the study found a clear negative correlation between universalism and support for policy areas typically associated with right-wing ideologies, such as border control, military spending, and law enforcement. Conversely, universalism was found to have a positive correlation with support for the four policy areas associated with left-wing ideologies.

Universalism proves to be a more significant factor in understanding individuals’ policy views and political ideology than traditional economic factors such as income, wealth, or beliefs about government efficiency and equity-efficiency preferences.

The findings of this study suggest a novel explanation for the formation of political ideologies in Western societies. The degree to which individuals exhibit the same level of altruism and trust towards out-group members as they do towards in-group members is a powerful predictor of their policy views and can help explain the grouping of individuals into distinct ideological clusters.

Peter Andre, briq postdoctoral fellow since January, has been selected as the winner of this year’s Distinguished CESifo Affiliate Award in the area of behavioral economics. The prize is funded by the international CESifo research network based in Munich and aims at identifying promising young researchers. Last year’s winner in the field behavioral economics was briq postdoc Suanna Oh (see briq newsroom article). Previous winners include briq research director Florian Zimmermann and briq research affiliate Matt Lowe.

Peter’s paper “Shallow Meritocracy” investigates whether people consider other people’s circumstances when judging their merits and accomplishments. Meritocracies aspire to reward hard work but promise not to judge individuals by the circumstances into which they were born. However, the choice to work hard is often shaped by individual chances and incentives resulting from one’s personal situation.

The award-winning study shows that people, in fact, often neglect personal circumstances and their effect on life choices when evaluating another person’s merits. In an experiment, US participants judge how much money workers deserve for the effort they exert. Unequal circumstances discourage some workers from working hard. Nonetheless, participants reward the effort of disadvantaged and advantaged workers identically, regardless of the circumstances under which choices are made.

Additional experiments identify an important underlying mechanism. Individuals understand that choices are influenced by circumstances. But, as people do not know what exactly would have happened on a level playing field, individuals base their merit judgments on the only reliable evidence they possess: observed effort levels.

See also the video interview: Three questions with Peter Andre