Document Type : Research Paper
Author
Faculty Member of Economics, Allameh Tabataba'i University, Tehran, Iran
Abstract
A substantial body of research highlights the presence of social preferences, their economic and political implications, and the varied conditions that influence their effects on the equilibrium and outcomes of human interactions. This study utilizes data from the Global Preferences Survey (GPS) to explore how age and gender impact social preferences across Iran at both national and provincial levels. The findings from a multivariate regression analysis reveal that age positively and significantly influences trust and altruism, while it negatively and significantly impacts negative reciprocity, with no significant effect on positive reciprocity. Gender, on the other hand, shows a significant negative effect on trust, indicating that women tend to exhibit lower levels of trust. Additionally, income and education negatively and significantly impact trust but positively affect altruism and positive reciprocity, with no notable effect on negative reciprocity.
The study also presents a provincial ranking within Iran based on key elements of social preferences: trust, altruism, positive reciprocity, and negative reciprocity. According to the research findings, Iran scores above the global average in terms of altruism, trust, and reciprocity. Notably, Markazi, Lorestan, and Ilam provinces rank highest in trust, altruism, and reciprocity, respectively.
Introduction
An influential set of laboratory experiments has questioned the conventional wisdom among economists and the validity of Stigler's (1981) position that "when self-interest and moral values conflict, most of the time, self-interest theory will win." These studies are complemented by a whole body of theoretical research that examines the nature and economic consequences of "social preferences". The most important of these theories consider social preferences as a result of altruism, reciprocity, fairness, and inequality-avoidance, maintaining social image or other motives. Studies based on field experiments also confirm the results of laboratory studies (List, 2006 and Falk et al., 2018). Ignoring social preferences makes economists unable to understand basic economic questions. Without considering social preferences, it is impossible to understand questions such as the effects of competition on market outcomes, the rules governing cooperation and collective action, the effects and determinants of material incentives, contracts and optimal property rights arrangements, and the important forces shaping social norms and market failure (Fehr and Fischbacher, 2002).
Determining factors of social preferences can be different according to demographic, geographic and cultural characteristics. Studies have shown that older people tend to prioritize fairness and reciprocation behavior (Falk et al., 2018). Also, studies have shown that women are typically more cooperative and tend to be more than men (Crawson & Gnezzi, 2009 and Falk & Hermel, 2018). Research has shown that older people and women typically have more trust (Crowson Vegnizi, 2009 and Duhman et al., 2008).
In Iran, due to the diversity of culture and geographical conditions, it is important to investigate the role of demographic variables on social preferences. In this research, we examined the causal relationship between age, gender, income, and education with each of the components of social preferences at the national and provincial levels with a standard model and technique.
Methods and Material
In order to investigate the factors affecting social preferences, the required data and information have been collected using a questionnaire. The information related to the dependent and independent variables of this research was extracted from the valid questionnaire available in Falk et al.'s article (2018) and the GPS website. Information related to Iran has been extracted from more than 2500 participants from all provinces of the country. In order to investigate the causal relationship between age and gender with variables of social preferences, multivariate regressions have been used in general as follows:
Results and Discussion
Using Ordinary Least Squares (OLS) regression, we estimated a model to examine the significance of gender, age, and cognitive ability on various dimensions of social preferences, including trust in others, altruism, and positive and negative reciprocity, as well as an aggregated index of social preferences. The analysis was conducted both at the national level and across individual provinces. Table 1 presents the estimated coefficients and their significance, highlighting how each variable influences these social preference metrics across different contexts within the country.
Table 1: Model estimation results for the whole country
independent variables
trust
altruism
Positive reciprocity
Negative reciprocity
Social preferences index
Dependent variables
age
0.0787***
0.006***
0.0053
-0.016**
0.0058*
Squared age
-0.0001
0.00005
-0.00004
-0.0003
-0.0002
gender
-0.1293***
0.049
0.0018
-0.0615
-0.1242
income
-0.045**
0.029***
0.03***
0.018
0.015**
education
-0.010***
0.022***
0.033***
0.012
0.014**
math
0.0165**
0.0095
0.0224***
0.0305***
0.0818***
R-Squared
0.0451***
0.007***
0.0068***
0.0582***
0.0126***
Sample size
2406
2474
2478
2449
2373
Source: Research calculations
*, **, *** are significant statistics at the level of 10, 5, and 1 percent, respectively.
The models were tested for heteroskedasticity, and where it was detected in the residuals, a robust estimator was applied. Specifically, in the models for positive reciprocity and the social preferences index, the null hypothesis of homoskedasticity was rejected, necessitating the use of robust estimations. Ramsey's test was conducted to check for potential endogeneity arising from omitted variables or specification errors, and this test did not reject the null hypothesis in any of the models. Table 2 provides the goodness-of-fit test results for these models.
Table 2: The results of goodness of fit tests
Tests/ Models
Trust
Altruism
Positive reciprocity
Negative reciprocity
Social preference index
Breusch–Pagan
(Prob)
1.88
(0.17)
1.75
(0.18)
11.65
(0.0006)
2.52
(0.11)
3.87
(0.0491)
Ramsey Test
(Prob)
1.03
(0.38)
0.26
(0.85)
0.15
(0.92)
1.81
(0.14)
1.25
0.29
Source: Research calculations
Conclusion
In this research, an attempt was made to investigate the effect of demographic variables such as age, gender, income, and education on social preferences for the entire country and province. The results of multivariate regression showed that, firstly, people's age has a positive and significant effect on trust in others and altruism. This finding is consistent with the findings of most studies. However, age has a negative and significant effect on negative reciprocity, which shows that negative reciprocity decreases with increasing age. And finally, age does not have a significant effect on positive reciprocity. Secondly, gender only has a negative and significant effect on trust in others and has no significant effect on other social variables. In other words, women are less devoted to others than men. The findings of this study are not consistent with the findings of most other studies on gender. Because in most other studies, women are more social, more altruistic and have significant negative countermeasures compared to men. The cause of this issue can be related to the culture and religion and the role of women in the country, which needs to be studied more in this field. Thirdly, income and education have a negative and significant effect on trust and have a significant and positive effect on altruism and positive reciprocity and do not have a significant effect on negative reciprocity. The effect of income and education is very similar. Income and education have a negative and significant effect on trust and have a significant and positive effect on altruism and positive reciprocity and do not have a significant effect on negative reciprocity.
Failure to understand and identify factors influencing social preferences leads to a misunderstanding of people's economic behavior (Fehr and Fischbacher, 2002). Correct understanding and identification of factors affecting social preferences helps policymakers to act optimally in understanding the process of cooperation between economic factors, designing economic incentives, and designing social policies.
Keywords
)