Do Institutions Affect Social Preferences? Evidence from Divided Korea

Byung‐Yeon Kim, Syngjoo Choi, Jungmin Lee, Sokbae Lee, Kyunghui ChoiJournal of Comparative Economics 2016 · DOI

TL;DR

This paper uses the unique "natural experiment" of divided Korea to investigate how institutions shape social preferences. By comparing North Korean refugees and South Korean students in economic games, they find that institutional experiences lead to distinct patterns of giving, with North Koreans exhibiting more egalitarian behavior. The study also highlights how attitudes towards market economy and democracy are linked to these preferences, though this connection is stronger for South Koreans.

What Problem Does This Paper Address?

Understanding the intricate relationship between institutions and individual preferences is a fundamental challenge in economics and social sciences. While it's widely believed that institutions shape our behavior and values, empirically disentangling this causal link is difficult because preferences and institutions often co-evolve. The authors leverage the division of Korea into two vastly different institutional systems (a market-oriented democracy in the South and a socialist dictatorship in the North) as a unique opportunity to study this question. They aim to determine if social preferences are endogenous to institutions and, if so, how these institutional differences manifest in observable behaviors and attitudes.

Methods

The researchers employed a combination of laboratory experiments and standard surveys.

  • Laboratory Experiments: They used widely recognized economic games, specifically variants of the dictator game (both "other-other" and "self-other" versions) and the trust game. These games are designed to elicit social preferences related to giving, trust, and reciprocity. A key innovation was informing participants about the group identity (North Korean refugee or South Korean student) of their anonymous partners, allowing for the examination of in-group versus out-group preferences.
  • Surveys: Standard surveys were administered to collect demographic information and measure subjective attitudes towards economic and political institutions, including support for market economy and democracy. For North Korean refugees, additional questions covered their experiences in North Korea and their resettlement in South Korea.

Datasets

The study utilized data collected from two distinct groups of participants in South Korea:

  • North Korean Refugees (NK): Two groups were recruited:
  • Study 1: Newly arrived refugees (within 1.5 years of the study).
  • Study 2: Refugees who had been in South Korea longer and were university students.
  • South Korean Students (SK): University students born in South Korea.

No pre-existing datasets were used for training or evaluation; the data was collected specifically for this research.

Key Insights and Ideas

  • Institutions leave a mark: The stark institutional differences between North and South Korea are reflected in the social preferences of their citizens, even after North Koreans have lived in the South for some time.
  • Egalitarianism under socialism: North Korean refugees, on average, demonstrate more egalitarian behavior in giving (dictator game) than South Korean students, suggesting that socialist institutions may foster a stronger preference for equal distribution.
  • Trust and reciprocity are less differentiated: While giving behavior shows clear differences, trust and reciprocity in the trust game do not exhibit significant group differences once preferences for giving are controlled for. This suggests that these specific social preferences might be more resilient or less directly shaped by the tested institutional differences.
  • Market economy attitudes matter more for South Koreans: The link between attitudes towards the market economy and experimental behavior (giving and trust) is much stronger and more consistent among South Korean subjects than among North Korean refugees. This implies that for South Koreans, their economic ideology more directly translates into their social behavior.
  • Democracy and trust connection: Support for democracy is positively correlated with trust, but this association is also significantly stronger for South Koreans.

Key Results

  • Dictator Game: North Korean subjects were significantly more egalitarian in their giving decisions, allocating a larger proportion of money to others compared to South Korean subjects. For instance, only 4% of North Koreans gave zero to their anonymous counterparts, versus 30% of South Koreans.
  • Trust Game: While North Koreans initially transferred more money, after controlling for giving preferences, there were no significant differences in trust between NK and SK subjects. However, South Koreans transferred more to North Korean opponents than to South Korean opponents, suggesting some level of sympathy.
  • Attitudes: North Korean subjects showed significantly less support for market economy and democracy compared to South Korean subjects.
  • Regression Analysis:
    • Giving behavior was negatively correlated with support for the market economy, meaning those favoring the market economy were less likely to give. This effect was stronger for South Koreans.
  • Trust was positively correlated with support for democracy, meaning those favoring democracy were more likely to trust. This effect was also stronger for South Koreans.
  • Reciprocity behavior showed no significant group differences and weak associations with institutional attitudes.

Limitations and Future Work

  • Selected Sample of Refugees: The study acknowledges that North Korean refugees are a selected sample, potentially more open to South Korean society. This might lead to an underestimation of the true differences between North and South Koreans.
  • Adaptation Challenges: The study notes that North Korean refugees face challenges in adapting to South Korean society due to a lack of education and skills. While the study controls for some of these factors, it's a complex issue.
  • Future Research: The authors suggest that understanding how social preferences evolve over time, especially for refugees assimilating into a new society, is a crucial area for future research. They also pose the question of whether these differences in social norms and preferences are important for the assimilation process.

Summary updated

BibTeX
@article{kim2016do,
  author = {Byung‐Yeon Kim and Syngjoo Choi and Jungmin Lee and Sokbae Lee and Kyunghui Choi},
  title = {{Do Institutions Affect Social Preferences? Evidence from Divided Korea}},
  journal = {Journal of Comparative Economics},
  year = {2016},
  doi = {10.1016/j.jce.2016.08.004},
}
Figure