Informal institution meets child development: Clan culture and child labor in China
Abstract
EconStor is a publication server for scholarly economic literature, provided as a non-commercial public service by the ZBW.
Full text
Tang, Can; Zhao, Zhong Working Paper Informal institution meets child development: Clan culture and child labor in China UNU-MERIT Working Papers, No. 2022-032 Provided in Cooperation with: Maastricht Economic and Social Research Institute on Innovation and Technology (UNU-MERIT), United Nations University (UNU) Suggested Citation: Tang, Can; Zhao, Zhong (2022) : Informal institution meets child development: Clan culture and child labor in China, UNU-MERIT Working Papers, No. 2022-032, United Nations University (UNU), Maastricht Economic and Social Research Institute on Innovation and Technology (UNU-MERIT), Maastricht This Version is available at: https://hdl.handle.net/10419/326841 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-nc-sa/4.0/
#2022-032 Informalinstitutionmeetschilddevelopment:Clancultureand childlaborinChina CanTangandZhongZhao Published10October2022 MaastrichtEconomicandsocialResearchinstituteonInnovationandTechnology(UNU‐MERIT) email:[email protected]u|website:http://www.merit.unu.edu Boschstraat24,6211AXMaastricht,TheNetherlands Tel:(31)(43)3884400
UNU-MERIT Working Papers ISSN 1871-9872 Maastricht Economic and social Research Institute on Innovation and Technology UNU-MERIT UNU-MERIT Working Papers intend to disseminate preliminary results of research carried out at UNU-MERIT to stimulate discussion on the issues raised.
Informal Institution Meets Child Development: Clan Culture and Child Labor in China Can Tang School of Business, Shanghai University of International Business and Economics Zhong Zhao※ School of Labor and Human Resources, Renmin University of China August, 2022 Abstract Using a national representative sample, the China Family Panel Studies, this paper explores the influences of clan culture, a hallmark of Chinese cultural history, on the prevalence of child labor in China. We find that clan culture significantly reduces the incidence of child labor and working hours of child laborer. The results exhibit strong boy bias, and are driven by boys rather than girls, which reflects the patrilineal nature of Chinese clan culture. Moreover, the impact is greater on boys from households with lower socioeconomic status, and in rural areas. Clan culture acts as a supplement to formal institutions: reduces the incidence of child labor through risk sharing and easing credit constraints, and helps form social norms to promote human capital investment. We also employ an instrument variable approach and carry out a series of robustness checks to further confirm the findings. Keywords: Informal institution; Clan culture; Child labor; China JEL codes: J22, J81, O15 We are grateful for very insightful discussions with and comments from Eric V. Edmonds and Douglas Staiger, and would like to thank constructive comments from the editor, two anonymous referees, and conference participants of the 7th International Workshop on Economic Analysis of Institutions organized by Xiamen University and seminar participants at Xiamen University, Central University of Finance and Economics, Kobe University and Shanghai University of Finance and Economics for their helpful comments. We also thank the Institute of Social Survey at Peking University for providing the China Family Panel Studies used in this paper. Can Tang acknowledges financial support from the Natural Science Foundation of China (Grant No. 72103130). Zhong Zhao acknowledges support by “Thematic Research Project on China's Income Distribution” (Project No. 21XNLG03) with funding from the research fund of Renmin University of China. ※ Corresponding author: Zhong Zhao, [email protected]. Zhong Zhao is also a member of GLO.
1 1 Introduction Child labor has many shortand long-term negative effects, such as reduction in human capital accumulation (Heady, 2003; Gunnarsson et al., 2006; Dumas, 2012; O’Donnell et al., 2005) and adverse effects on the future labor market outcomes (Emerson and Souza, 2011). Obviously, the welfare at both individual level and national level will be impeded by such negative impacts. Understanding the factors behind child labor is an important topic in the literature. In this paper, we focus on the role of clan culture on child labor in China. Our study shows that clan culture acts as a supplement to formal financial market by reducing the incidence of child labor through risk sharing and easing credit constraints, which enriches the literature on the macro factors behind the incidence of child labor, such as the role of public policies, globalization and financial development (Dehejia and Gatti, 2005; Edmonds and Pavcnika, 2005; Dinopoulos and Zhao, 2007; Tang et al., 2020; Zhao et al., 2021). An imperfect financial market in developing countries leads to high levels of child labor engagement for two reasons (Baland and Robinson, 2000; Ranjan, 2001; Udry, 2003). Firstly, the benefits of child labor come from the current wage earned by the child and the reduced cost of schooling. However, the cost of child labor is the future labor earnings impeded by involved in child labor. When facing credit constraints, a higher interest rate in an imperfect financial market leads to a lower present discounted value of the future labor earnings. Thus, the benefits would be higher relative to the cost, resulting in higher incidence of child labor (Udry, 2003). Secondly, decisions regarding child labor and schooling are generally made by parents, which raise the issue of agent problems. Agent problems become even more salient when they occur within imperfect financial markets (Udry, 2003). Even amongst altruistic parents, the proportion of child labor can produce a socially inefficient equilibrium, as imperfect financial markets prevent parents from fully internalizing negative effects (Baland and Robinson, 2000; Udry, 2003). Besides, as a part of the household’s self-insurance
2 strategy, child labor supply will increase when household facing negative shocks, especially when the insurance market is not well developed (Jacoby and Skoufias ,1996; Beegle et al., 2006; De Janvry et al., 2006). Culture is a vital informal institution, and is identified in emerging economies as a crucial determinant of many economic outcomes, especially when the formal institution is poorly developed (Guiso et al., 2006; Fernández, 2011; Voigtländer and Voth, 2012; Alesina and Giuliano, 2015). However, there are few economic studies focusing on the impact of culture on child labor. Thus, our study also contributes to the growing literature of cultural economics by examining the impact of clan culture on the incidence of child labor in China. We focus on China for three reasons. Firstly, although the enrollment rates of primary and junior high schools are high in China 1 , approximately 7.74% of children aged between 10 and 15 years were engaged in child labor in 2010 (Tang et al., 2018). It is crucial to investigate how to reduce the child labor participation in China. Secondly, clan culture is a well-known hallmark of Chinese culture (Greif and Tabellini, 2010, 2017; Zhang, 2020). Whilst official clan organizations were largely abolished following the Chinese Communist Party’s rise to power in 1949, the influences of clan culture are likely to persist and continue to have an important impact on the behavior of individuals in modern China (Greif and Tabellini, 2010, 2017). Thirdly, China is the largest transition and developing economy, but has yet to establish a well-developed financial market, legal system, and social safety system, particularly in rural areas. In such context, clan culture plays the role of formal institutions in promoting information and risk sharing, reducing transaction costs, overcoming financial constraints for enterprises, and easing credit constraints for its members (Weber, 1981; Peng, 2004; Greif and Tabellini, 2010; Cox and Fafchamps, 2007; Zhang, 2020). Besides, clans also provide public goods and services for clan members (Fei, 1946; Freedman, 1965; Tsai, 2007; Gerard et al., 2015). 1 The enrollment rate of children aged between 10 to 15 years is 97% in 2010, according to our calculation based on CFPS 2010.
3 Combing the theories of child labor determination and the functions of clan culture, we hypothesize that clan culture can influence the engagement of child labor. Clan culture, as a supplement to formal institutions, may reduce the labor supply of children from poor families by easing credit constraints and promoting risk sharing. Moreover, this impact is supposed to be biased toward boys due to the patrilineal clan nature. To test our hypotheses, we use the indicator of whether a household has a genealogy book as a key measure of the strength of clan culture at the household level following existing literature (Peng, 2004; Chen et al., 2016; Greif and Tabellini, 2017; Zhang, 2020). Having a genealogy book means the household subscribing to a clan. Thus, this paper focus on the differences between clan members and non-clan members, other than the impacts of clan culture to regional outcomes or all individuals. That is, rather than exploring the role of external cultural environment (Guiso et al., 2004), we pay more attention to the intergenerational cultural transmission of preferences and beliefs within families (Tabellini, 2008), social norm among clan members, and the effect of being a clan member. Besides using a household level indicator, we also explore to measure the strength of clan culture by the surname concentration in the village following Peng (2004). Employing our two measures of clan culture, we attempt to separately identify the potential effects of external origins and family origins. On the one hand, concentration of the same surname reflects the strength of clan culture in the village, regardless of household norms. On the other hand, the genealogy book measures the importance of clan culture to a particular household, after controlling the village level indicator of clan culture. Based on national representative data, the China Family Panel Studies, we find that clan culture can significantly reduce the incidence of child labor. The results are driven by boys rather than girls. The differential effects between boys and girls indicate a strong engrained bias towards boys within the clan culture since its patrilineal nature, which influences parents to allocate resources towards boys. Moreover, the impact is
4 greater on boys from households with lower socioeconomic status, and in rural areas. We investigate risk sharing, informal financial institution, social norm on promoting education three mechanisms through which clan culture influences the incidence of child labor. Specifically, clan culture has a greater impact on reducing child labor engagement amongst boys when serious disasters occur. Additionally, clans are more likely to help fathers of boys, if they fall into difficulty. Finally, we provide evidence of how clan culture may increase study time and household educational expenditure on boys. Our paper contributes to existing literature in several ways. Firstly, it enriches the literature of exploring the impacts of culture on economic development. As existing literature shows, culture can effectively influence labor participation decision, fertility decision, schooling decisions, academic performance, wage rate, and labor mobility (e.g., Fernández, 2007; Guiso et al, 2008; Fernández and Fogli, 2009; Giavazzi et al., 2013; Munshi and Rosenzweig, 2006; Pitt, Rosenzweig and Hassan, 2012). However, seldomly do they focus on child labor participation. We emphasize that culture can also influence human capital accumulation and economic development through the child labor channel. Secondly, we contribute to existing child labor determination literature (Dehejia and Gatti, 2005; Edmonds and Pavcnika, 2005; Basu et al., 2010; Tang et al., 2018; Tang et al. 2020; Zhao et al., 2021) by identifying a new influencing factor: clan culture. Since formal institutions are weak in most developing countries, exploring the role of culture in child labor provides a new insight to understanding the determination and distribution of child labor. Thirdly, this paper highlights the potential threat associated with reliance on informal institutions and culture, such as clans, to fill the void of under-developed formal institutions, since the informal institution and culture can be gender biased. The remainder of the paper is organized as follows: Section 2 briefly introduces clan culture in China; Section 3 describes the data and outlines the empirical strategy;
5 Section 4 presents and discusses the main results; Section 5 provides robustness checks; Section 6 conducts heterogeneity analyses; Section 7 explores potential mechanisms and possible interpretations of the main findings; Section 8 provides concluding comments. 2 Clan Culture in China and Related Literature Clan culture is a hallmark culture in China (Greif and Tabellini, 2010, 2017; Zhang, 2020). The clans are kinship-based patrilineal organizations made up of individuals with one common male ancestor. The earliest clan organizations appeared within the Zhou Dynasty (1046-256 BC) and they prevailed within the Song Dynasty (960-1279 AD) when large conflicts broke out frequently in northern China, which led to great population migration to the south; the present spatial distribution of clans was mainly formed at that time (Feng, 2013). Zhu Xi (1130–1200) was a fervent advocate of clans and provided detailed institutional designs for Chinese clan organizations, which became the standard social practices in the subsequent eight centuries (Cheng et al., 2021). For management convenience and stronger cohesiveness, a traditional clan typically had an elder board to administer clan affairs, an ancestral temple to offer sacrifices to ancestors, a genealogy book recording brief biographies of all descents from the apical ancestor, and a code of conduct to direct and constrain clan members. Besides, a typical clan usually has common properties that can be used to assist members in need. Chinese clans have two basic features. Firstly, clans appreciate family values, obligations and loyalty (Fei, 1946; Liu, 1959, Peng, 2004). Due to the sense of duty, clan members are more inclined to make clan-oriented decisions. In return, clans also provide protection for loyal members. However, clan culture leads to strong boy preference. As cooperation organizations consisting of patrilineal households that trace their origin to a common male ancestor, clans emphasize patrilineal descent so that only males can pass on their family names to next generations and have the right to be clan leaders (Feng, 2013). Along with clan culture, the Chinese family system is patriarchal,
12 All models are estimated using the sample weights provided by the CFPS. In all regressions, standard errors are adjusted for clustering at the household level to account for any potential correlation within the same household (Bertrand et al., 2004) 9 . Although we only have cross-sectional data, culture is usually regarded as exogenous. The cultural traits transmit from generation to generation (Grief, 1994; Guiso et al., 2006), and influence economic outcomes through two potential channels: beliefs (i.e., priors) and values (i.e., preferences), which can only be reshaped in the long term (Guiso et al., 2006). Thus, culture can be regarded as persistent and stable. According to Becker (1996, p. 16): “Individuals have less control over their culture than over other social capital. They cannot alter their ethnicity, race or family history, and only with difficulty can they change their country or religion. Because of the difficulty of changing culture and its low depreciation rate, culture is largely a ‘given’ to individuals throughout their lifetimes.” Furthermore, whether a household holds a genealogy book is relatively exogenous to the household. A genealogy book usually records the history of a clan, including brief biographies of all descents from the apical ancestor and descriptions of all mega events and honorable descents in the clan. Compiling a genealogy requires detailed information on all extended-family members including previous generations and current generation. Therefore, whether a household holds a genealogy book is mostly determined by the extended-family other than an individual household. 4 Main Results In this section, we firstly present results from the baseline model, and then deal with the potential endogeneity issues. The main outcomes of interest are child labor dummy variable and sum of paid and unpaid of working hours of children. 9 We also consider clustering standard error at the village/community level as robustness checks. The main empirical results are still hold.
13 4.1 Baseline results Table 4 reports the baseline results. All regressions in Table 4 include individual characteristics, household characteristics, village/community characteristics, and county fixed effects. The estimated coefficients of clan culture strength are negative and statistically significant in Column (1), suggesting that clan culture helps combat child labor. Particularly, a child from a clan household is about 2.7% less likely to be a child laborer. Since clans in China emphasize patrilineal descent, clan culture leads to a strong boy preference. Thus, we further explore whether clan culture has differential effects on boys and girls. The results are shown in Columns (2)-(4) in Table 4. We find that all the effects go through the interaction term of genealogy and boy; the coefficients of genealogy become insignificant in Column (2). In other words, clan culture helps combat child labor only for boys. We run separate regressions to quantify the differential impact on boys and girls, with the results for girls in Column (3) and boys in Column (4). Results show that clan culture significantly helps reduce the incidence of child labor for boys. However, it does not have a significant effect for girls. [Table 4 about here] Since there is no significant difference between the overall likelihood of child labor for boys and girls in China (Tang et al., 2018), 10 the stronger impact of clan culture on boys may not come from the different level of child labor incidence between boys and girls. When a family is faced with severe credit constraints, both sons and daughters may have to engage in economic activities. Clan culture may relax household credit constraints and lead to intra-household resource reallocation towards boys due to strong boy preference within clans (we will verify this hereinafter). Consequently, a relaxation of resource constraints by clans may create differential effects for boys and girls. 10 In the sample used in our analyses, 7.6% of girls and 7.9% of boys engaged in child labor.
14 Besides economic activities, children also involve non-economic housework, such as taking care of the family and household chores. We also examine the children’s noneconomic housework participation. According to our calculation from the CFPS 2010, 51.34% of children participate in housework. On average, child laborers spend 6.72 hours per day on individual work on average, whilst they only spend 0.847 hours per day on housework. When we consider the impacts of clan culture on children’s housework participation, we find no significant effects. In other words, clan culture helps reduce children’s participation in economic activities, but has no impacts on their housework participation. 11 4.2 Correcting for the potential endogenous bias The endogeneity concerns may come from three aspects in this research: reverse causality, measurement error of clan culture, and omitted variables. Since whether a household has a genealogy book is hard to be influenced by child labor engagement, reverse causality issue is not the main concern in our specification. As for measurement error of clan culture, it is possible for a household emphasizing clan culture do not keep a genealogy book, but seldomly would a non-clan household keep a genealogy book. Measurement error in this direction underestimates the impact of clan culture. Therefore, the major endogeneity concerns may come from omitted variables. To assess the importance of potential omitted variable bias, we perform a test proposed by Oster (2019). This method analyzes the differences in coefficients of interest and the R-squared of the main regressions between excluding controls and including controls. If including controls increases the R-squared of the model while not affecting the coefficient of interest, it is less likely that including unobservables would bias the results. Oster (2019) assesses this potential bias by compute the relative importance of unobservables to observables (𝛿) that would be consistent with a 11 The results are available from the authors upon request.
15 coefficient of interest equal to zero (𝛽=0). That is, how important should the unobservables to be relative to the observables to eliminate the estimated effect. For our baseline specification, the 𝛿 that matches 𝛽=0 amounts to 2.208 for whole sample, 3.629 for girls and 1.871 for boys. It means that the importance of unobservables would have to be 2.208, 3.629 and 1.871 times higher than that of the observables for the coefficients to be zero. According to Oster (2019), the omitted variable bias would be less of a concern if the estimated 𝛿 exceeds one. The values of 𝛿 are far more than one in our tests, indicating that omitted variables bias is limited. To further control for the possible endogenous issue and assess the importance of the potential endogenous bias, we adopt an instrument variable (IV) approach and carry out a Hausman test. Referring to Chen et al. (2021), we instrument clan culture with a city’s shortest distance to the academic center where Zhu Xi (1130–1200 CE) developed and spread his philosophy. Zhu Xi was a fervent advocate of clans and provided detailed institutional designs for Chinese clan organizations, which became the standard social practices in the subsequent eight centuries (Cheng et al., 2021). Zhu had set up 3 academies during his whole life, all of which were located in Nanping (in Fujian province) and the majority of Zhu’s classical texts were completed there, such as Jiali (the Family Rituals) (Chan, 1987). In Jiali, Zhu not only elaborated on the importance of establishing clan common properties, such as ancestral halls, ancestral graveyards, and ritual land, but also laid out detailed instructions on the design of ancestral halls and the rules for family rituals like adult ceremonies, marriages, funerals, and ancestral offerings etc. (Cheng et al. 2021). Meanwhile, of the three academies Zhu built, the most important one is “Kaoting Academy”, because most of his famous pupils attended him here from all over China and most of the recorded conversations took place here, forming the “Kaoting School” with great influence in history (Chan, 1987). In a word, Nanping was the academic center of spreading Zhu Xi’s main ideas about clan culture. Thus, we expect that regions closer to Nanping were more exposed to Zhu Xi’s influences historically and, thanks to
16 cultural persistence, are more likely to preserve strong clan culture heritage today. Besides, the location choice of the academic center where Zhu Xi (1130–1200 CE) developed and spread his philosophy is largely random. Zhu Xi chose to establish academies in Nanping because his father used to be a local government official there and his mother was buried there. Nanping was neither an economic center nor backward regions. Therefore, consistent with Chen et al. (2021) and Cheng et al. (2021), we argue that the IV does not affect the child labor decision through other channels than clan culture. So, it is reasonable to use a city’s shortest distance to Nanping as an IV of the strength of clan culture. The instrumental variable (IV) results are shown in the Table 5. These results reinforce our baseline conclusion: the negative effects of clan culture on the incidence of child labor are significant for the full sample and the boy sample. [Table 5 about here] Furthermore, we employ a Hausman specification test to examine whether it is necessary to adopt an IV approach (Hausman, 1978). The result from a robust Hausman test using bootstrap with 1500 replications shows Prob > chi2=0.23, which means that we cannot reject the null hypothesis that all explanatory variables are exogenous. Since OLS is more efficient than IV approach if all explanatory variables are exogenous, and both test of Oster (2019) and Hausman test indicate that there is no significant endogenous bias, it is rational for us to choose OLS method as the main empirical strategy in this paper 12 . 12 It should be added that, in the OLS regressions we control for the county fixed effect. When it comes to the IV approach, since our instrument variable - a city’s distance to Nanping - is a part of the county fixed effect, we cannot control for the county fixed effect anymore due to perfect collinearity. Thus, to execute the Hausman specification test, the regression equation of OLS we use here is the same as the regression equation of IV approach (without controlling for county fixed effect). According to Mundlak (1987), we can always find a linear projection to transfer a fixed effect model to a random effect model, so it is rational to transfer the OLS regression equation controlling for county fixed effect to an equation without controlling for county fixed effect.
17 4.3 Working hours In this part, we go one step further to examine the effects of clan culture on working hours of children. Since a substantial proportion of children are observed with zero working hour, we use a tobit model. Results in Table 6 show that clan culture significantly helps reduce children’s working hours, and again this result is driven by boys other than girls. [Table 6 about here] Because a majority of children are not working, it is important and interest to investigate how the clan culture effects on the extensive margin (the participation of in the labor market) and on the intensive margin (working hours conditional on working) by McDonald and Moffitt's (1980) decomposition. Table 7 reports the decomposition results. Panel A and B show results for the whole sample, while Panel C for girls and D for boys. Clan culture has reduced both the extensive and the intensive margin, but the effects are again significant for boys only. Panel D illustrates that clan culture reduces labor participation of boy by 2.6% at extensive margin and reduces working time by 0.24 hours at intensive margin. [Table 7 about here] 5 Robustness checks In this subsection, we conduct a series of robustness checks for our baseline results: permutation test; sample restriction; alternative measure of the strength of clan culture; taking north-south difference into account; and consideration of different ethnic groups. 5.1 Permutation test We adopt a permutation test of Rosenbaum (2007) to check whether the estimated results are really significant or just due to random chance. The permutation
18 test makes no assumptions on the underlying distribution of the data. In the permutation test, the null hypothesis is that the clan culture has no effect on the odds of child labor. Under the null hypothesis, the estimated coefficient from the actual data can be considered a random sample from the permutation distribution. We can produce the permutation distribution of the estimated coefficients and use it for statistical inference. We randomly assign whether or not the family household has a genealogy book as the placebo treatment status for each child (Rosenbaum, 2007; Lu and Anderson, 2015). Specifically, the ratio of households having genealogy books in placebo treatment status is consistent with that in the actual data. We estimate the placebo treatment effect on the incidence of child labor. The distribution of placebo treatment effects from 4000 random assignments are displayed in Fig.2. The dashed line shows the estimated treatment effect from the baseline analysis. The p value of the permutation placebo test is the proportion of placebo estimates that are equal to or larger in absolute value than the corresponding estimate from the baseline analysis. We find that the p value is 0.033, which rejects the null hypothesis of no effect, thus providing further support for our identification strategy and main findings. [Figure 2 about here] 5.2 Sample restriction Clans emphasize patrilineal descent and have a strong boy preference. Thus, when we divide the whole sample into the subsamples of boys and girls, the classification may by endogenous. As families with sons may be more deeply influenced by clan culture, i.e., through fertility decisions, it is possible that the differential gender effects shown in the baseline results may not only capture the impact of clan culture on the incidence of child labor, but also include different strengths of clan culture between these two subsamples. As a robustness check, we restrict our analysis to the children from households with both sons and daughters. On the one hand, such families are more likely to have the same gender preference, but on the other hand, intra-household resource allocation between boys and girls is more relevant to families with both sons
19 and daughters. The results are in Table 8: clan culture still significantly reduces the incidence of child labor for boys, but not for girls, and the differential gender effects are indeed stronger for such families. 13 [Table 8 about here] 5.3 Alternative measures of the strength of clan culture We then check if our results are sensitive to alternative measures of clan culture strength. Where clan culture is prevalent, extended family members tend to live in the same village. In reference to Peng (2004), we measure the strength of clan culture by the surname concentration in the village. Particularly, as shown in Columns (1)-(3) in Table 9, we adopt whether or not there is a surname shared by over 10% of village households as an alternative measure of the strength of clan culture. 14 We exclude the urban sample in this part, as we do not have any information about surname prevalence at community level within urban areas. Using the alternative measure of clan culture strength does not alter our main finding: clan culture has a negative and significant effect on the incidence of child labor for boys. Furthermore, some literature focuses on the intergenerational cultural transmission of preferences and beliefs within families (Tabellini, 2008), whilst others shed light on the role of external cultural environment (Guiso et al., 2004). Thus, an interesting question to consider is whether it is intergenerationally transmitted clan culture within families, or the area in which an individual lives that helps combat child labor. Employing our two measures of clan culture, we attempt to separately identify the potential effects of external origins and family origins. On the one hand, concentration of the same surname reflects the strength of clan culture in the village, regardless of household norms. On the other hand, the genealogy book measures the importance of 13 We also consider the impacts of interprovincial migrants to our results. Only 1.4% of our sample are interprovincial migrants. When we restrict our sample to local children, our main results do not change much. The results are available from the authors upon request. 14 This indicator comes from a question of CFPS2010 in village/community module, which ask “In your village, was there any surname shared by over 10% of all the households?”. There is no additional information available to use other cutoff.
20 clan culture to a particular household, after controlling the village level indicator of clan culture. Specifically, we add clan culture measures of both village level and household level to the regressions and report the results in Columns (4)-(6) of Table 9. Interestingly, both indicators are significantly negative in the regression for boys. That is, clan culture does help reduce child labor participation of boys irrespective of whether it is transmitted within the family or prevalent in the village. [Table 9 about here] 5.4 Consider difference in clan culture among ethnic groups There are 56 ethnic groups in China, including the Han Chinese accounting for about 91.51% of the total population and rest are 55 other ethnic minorities. 15 Since different ethnic groups may hold different beliefs, we additionally check for robustness across different ethnic groups by controlling the interaction of the strength of clan culture and whether the village/residential community is a minority-concentrated area. The results are reported in Table 10. The estimated results are quite close to those obtained from baseline results. The coefficients of the dummy variable on whether a household belongs to Han Chinese, if a village/community is a minority-concentrated area, and the interaction term of clan culture and minority-concentrated area are all insignificant, which implies that clan culture plays an important role in reducing child labor amongst both the Han Chinese and minorities. [Table 10 about here] 5.5 Take north-south difference into account Figure 1 shows clan culture concentrates in southern regions of China. The uneven distribution may imply different impacts of clan culture on child labor 15 Data source: 2010 National Census of China.
21 participation between northern and southern provinces. As shown in Table 11, columns (1)-(3) show the estimates for children living in northern provinces whilst the estimates for children living in southern provinces are reported in Columns (4)- (6). We find that clan culture significantly reduces the incidence of child labor for children in southern provinces, especially for boys, but has no significant effect on the engagement of child labor for children in northern provinces. 16 [Table 11 about here] 6 Mechanisms and Interpretations 6.1 A supplement to formal institutions Since culture acts as a supplement to formal institutions, especially when the formal institution is not well developed (Alesina and Giuliano, 2015), the impact of clan culture on children’s labor force participation may vary between urban and rural areas due to the different development of formal institution and economic level. We investigate rural-urban heterogeneity to test whether clan culture acts as a supplement to formal institutions and plays a much important role in rural China. As shown in Table 12, Panel A presents the effects of clan culture by rural/urban hukou status, while Panel B refers to the effects of clan culture by rural/urban residence. Columns (1)-(3) show the estimates for urban children whilst the estimates for rural children with are reported in Columns (4)-(6). The results show that clan culture helps reduce the incidence of child labor for rural children, especially for rural boys, but has no significant effect on the likelihood of child labor amongst both urban girls and urban boys. [Table 12 about here] 16 Northern China includes Beijing, Tianjin, Hebei, Shanxi, Inner Mongolia, Liaoning, Jilin, Heilongjiang, Shandong, Henan, Shaanxi, Gansu, Qinghai, Ningxia and Xinjiang. Southern China includes Shanghai, Jiangsu, Zhejiang, Anhui, Fujian, Jiangxi, Hubei, Hunan, Guangdong, Guangxi, Hainan, Chongqing, Sichuan, Guizhou, Yunnan, and Tibet.
28 Cox, D., Fafchamps, M., 2007. Chapter 58 Extended Family and Kinship Networks: Economic Insights and Evolutionary Directions. Elsevier B.V. de Janvry, A., Finan, F., Sadoulet, E., Vakis, R., 2006. Can conditional cash transfers serve as safety nets in keeping children at school and from working when exposed to shocks? Journal of Development Economics 79, 349–373. Dehejia, Rajeev H., Gatti, R., 2005. Child Labor: The Role of Financial Development and Income Variability across Countries. Economic Development & Cultural Change 53, 913-931. Dinopoulos, E., Zhao, L., 2007. Child Labor and Globalization. Journal of Labor Economics 25, 553-579. Dumas, C., 2012. Does Work Impede Child Learning? The Case of Senegal. Economic Development & Cultural Change 60, 773-793. Edmonds, E. V., 2008. Child Labor. Handbook of Development Economics, 4, 36073709 Edmonds, E.V., Pavcnik, N., 2005. The effect of trade liberalization on child labor. Journal of International Economics 65, 401-419. Emerson, P., Souza, A., 2007. Child labor, school attendance and intra-household gender bias in Brazil. World Bank Economic Review 21, 301–316. Emerson, P.M., Souza, A.P., 2011. Is Child Labor Harmful? The Impact of Working Earlier in Life on Adult Earnings. Economic Development and Cultural Change 59, 345-385. Fei, H.T., 1946. Peasantry and Gentry: An Interpretation of Chinese Social Structure and Its Changes. American Journal of Sociology 52, 1-17. Feng, E., 2013. Zhongguo Gudai de Zongzu he Citang [clans and Ancetral temples in ancient China]. The Commercial Press. Fernández, R., 2007. Alfred Marshall Lecture: Women, Work, and Culture. Journal of the European Economic Association 5, 305-332. Fernández, R., 2011. Chapter 11 - Does Culture Matter? in: Benhabib, J., Bisin, A., Jackson, M.O. (Eds.), Handbook of Social Economics, Vol. 1. North-Holland, pp. 481-510. Fernández, R., Fogli, A., 2009. Culture: An Empirical Investigation of Beliefs, Work, and Fertility. American Economic Journal Macroeconomics 1, 146-177. Freedman, B.M., 1965. Lineage organization in southeastern China. Pacific Affairs 32, 207. Gerard, P.I.M., Qian, N., Xu, Y., Yao, Y., 2015. Making Democracy Work: Culture, Social Capital and Elections in China. Social Science Electronic Publishing.
29 Giavazzi, F., Schiantarelli, FM., Serafinelli, MF., 2013. Attitudes, Policies, and Work attitudes, polices, and work. Journal of the European Economic Association 11, 1256–1289. Greif, A., 1994. Cultural Beliefs and the Organization of Society: A Historical and Theoretical Reflection on Collectivist and Individualist Societies. Journal of Political Economy 102, 912-950. Greif, A., Tabellini, G., 2010. Cultural and Institutional Bifurcation: China and Europe Compared. American Economic Review 100, 135-140. Greif, A., Tabellini, G., 2017. The clan and the corporation: Sustaining cooperation in China and Europe. Journal of Comparative Economics 45, 1-35. Guiso, L., Monte, F., Sapienza, P., Zingales, L., 2008. Diversity. Culture, gender, and math. Science 320, 1164. Guiso, L., Sapienza, P., Zingales, L., 2004. The Role of Social Capital in Financial Development. American Economic Review 94, 526-556. Guiso, L., Sapienza, P., Zingales, L., 2006. Does Culture Affect Economic Outcomes? Journal of Economic Perspectives 20, 23-48. Gunnarsson, V., Orazem, P.F., Sánchez, M.A., 2006. Child Labor and School Achievement in Latin America. World Bank Economic Review 20, 31-54. Harrell, S., 2002. Patriliny, patriarchy, patrimony surface features and deep structures in the Chinese family system. Paper draft. URL http://faculty.washington.edu/stevehar/PPP.html Hausman, J. A., 1978. Specification Tests in Econometrics. Econometrica, 46, 12511271. Heady, C., 2003. The Effect of Child Labor on Learning Achievement. World Development 31, 385-398. Jacoby, H., Skoufias, E., 1996. Risk, financial markets, and human capital in a developing country. Review of Economic Studies 64, 311–335. Knodel, J., Wongsith, M., 1991. Family size and children’s education in Thailand: Evidence from a national sample. Demography 28, 119–131. Liu, H. W., 1959. An analysis of Chinese clan rules: Confucian theories in action. Stanford: Stanford University Press. Lu, F, Anderson M., 2015. Peer Effects in Microenvironments: the Benefits of Homogeneous Classroom Groups. Journal of Labor Economics 33, 91–122. McDonald, J. F., Moffitt, R., 1980. The uses of Tobit analysis. The Review of Economics and Statistics, 62, 318–321.
30 Moehling, C., 2004. Family structure, school attendance, and child labor in American south in 1900 and 1910. Explorations in Economic History 41, 73–100. Mundlak, Y., 1978. On the pooling of time series and cross section data. Econometrica, 46(1), 69-85. Munshi, K., Rosenzweig, M., 2006. Traditional Institutions Meet the Modern World: Caste, Gender, and Schooling Choice in a Globalizing Economy. The American Economic Review, 96, 1225–1252 Nunn, N., Wantchekon, L., 2011. The Slave Trade and the Origins of Mistrust in Africa. American Economic Review 101, 3221-3252. O'Donnell, O., Rosati, F.C., Doorslaer, E.V., 2005. Health effects of child work: Evidence from rural Vietnam. Journal of Population Economics 18, 437-467. Oster, E., 2019. Unobservable selection and coefficient stability: Theory and evidence. Journal of Business & Economic Statistics, 37, 187–204. Peng, Y., 2004. Kinship Networks and Entrepreneurs in China’s Transitional Economy1. American Journal of Sociology 109, 1045-1074. Pitt, M. M., Rosenzweig, M. R., Hassan, N., 2012. Human Capital Investment and the Gender Division of Labor in a Brawn-Based Economy. The American Ranjan, P., 2001. Credit constraints and the phenomenon of child labor. Journal of Development Economics 64, 81-102. Rosenbaum, PR., 2007. Interference between Units in Randomized Experiments. Journal American Statistical Association 102, 191–200. Tabellini, G., 2008. The Scope of Cooperation: Values and Incentives. Quarterly Journal of Economics 123, 905-950. Tang, C., Zhao, L., Zhao, Z., 2018. Child Labor in China. China Economic Review 51, 149-166. Tang, C., Zhao, L., Zhao, Z., 2020. Does Free Education Help Combat Child Labor? The Effect of a Free Compulsory Education Reform in Rural China. Journal of Population Economics 33, 601-631. Tsai, L.L., 2007. Solidary Groups, Informal Accountability, and Local Public Goods Provision in Rural China. American Political Science Review 101, 355-372. Udry, C.R., 2003. Child Labor. Working Papers 3, 79-79. Voigtländer, N., Voth, H.J., 2012. Persecution Perpetuated: The Medieval Origins of Anti-Semitic Violence in Nazi Germany*. Quarterly Journal of Economics 127, 1339-1392. Wahba, J., 2006. The influence of market wages and parental history on child labour and schooling in Egypt. Journal of Population Economics 19, 823–852.
31 Watson, J.L., 1982. Chinese Kinship Reconsidered: Anthropological Perspectives on Historical Research. China Quarterly 92, 589-622. Weber, M., 1981. General Economic History. New Brunswick, N.J.: Transaction Books. Xie, Y., 2012. China family panel studies (2010) User's manual. Institute of Social Science Survey: Peking University, Beijing, China. Xu, Y., Yao, Y., 2015. Informal Institutions, Collective Action, and Public Investment in Rural China. American Political Science Review 109, 371-391. Young HP., 2015. The evolution of social norms. The Annual Review of Economics 7, 359–87. Zhang, C., 2019. Family Support or Social Support? The Role of Clan Culture. Journal of Population Economics 32, 529-549. Zhang, C., 2020. Clans, Entrepreneurship, and Development of the Private Sector in China. Journal of Comparative Economics 48, 100-123. Zhao, L., Wang, F., Zhao, Z., 2021. Trade Liberalization and Child Labor. China Economic Review 65.
32 Figure 1. Geographical Distribution of Clan Culture Notes: 1. This figure shows the proportion of households holding genealogy books at province level; 2. Authors’ calculation based on the family questionnaire of CFPS 2010.
33 Figure 2. Child Labor Engagement of Boys, p-value=0.033 Note: Estimated coefficients from the permutation placebo tests. We randomly assign whether or not there is a genealogy book to the children’s households as placebo treatment status for the same sample in the baseline analysis. These histograms display the distribution of placebo treatment effects from 4000 random assignments. The dashed line shows the estimated treatment effect from the baseline analysis. The p value of each permutation placebo test is the proportion of placebo estimates that are equal to or larger in absolute value than the corresponding estimate from the baseline analysis.
34 Table 1. Strength of Consanguinity Ties Variables （1） （2） （3） （4） （5） Households without genealogies Households with genealogies Difference Obs Mean Obs Mean Participate in ancestor worship/tomb-sweeping activities last year (dummy) 11378 0.676 3322 0.766 -0.090*** How often did the family do the following contacts with relatives and friends last month ： Entertainment/ dine together 11369 0.575 3319 0.723 -0.148*** Give food or gifts 11374 0.381 3322 0.568 -0.187*** Help each other 11371 0.384 3319 0.497 -0.114*** Pay a visit 11368 0.742 3320 0.888 -0.146*** Chat 11367 1.147 3321 1.287 -0.141*** Notes: 1. Frequencies of contacts are divided into 5 categories: 0, no contact at all; 1, once a month; 2, 2-3 times a month; 3, 2-3 times a week; 4, almost every day. 2. Authors’ calculation based on the family questionnaire of CFPS2010. 3. T-test of difference between households with a genealogy book and without is reported in Column (5). 4. Results are weighted using sample weights provided in the data.5. *** p<0.01, ** p<0.05, * p<0.10.
35 Table 2. Summary Statistics (1) (2) (3) (4) (5) Variables Obs Mean Std. Dev. Min Max Child labor (dummy) 3435 0.078 0.268 0 1 Genealogy (dummy) 3398 0.289 0.453 0 1 Age 3435 12.473 1.711 10 15 Han nationality (dummy) 3435 0.835 0.371 0 1 Boy (dummy) 3435 0.537 0.499 0 1 Rural Hukou (dummy) 3405 0.772 0.419 0 1 Interprovincial migrant (dummy) 3426 0.014 0.116 0 1 Education of father (year) 3356 7.015 4.288 0 19 Education of father (category) Illiteracy 3382 0.171 0.376 0 1 Primary School 3382 0.271 0.445 0 1 Junior High School 3382 0.374 0.484 0 1 Senior High School 3382 0.124 0.33 0 1 College and above 3382 0.06 0.238 0 1 Education of mother (year) 3357 5.625 4.563 0 22 Education of mother (category) Illiteracy 3386 0.3 0.458 0 1 Primary School 3386 0.273 0.446 0 1 Junior High School 3386 0.298 0.458 0 1 Senior High School 3386 0.083 0.276 0 1 College and above 3386 0.045 0.208 0 1 Age of father 3404 40.257 4.995 28 75 Age of mother 3389 38.375 4.619 24 72 Household head is male (dummy) 3,426 0.758 0.428 0 1 Number of children 3431 1.824 0.926 1 7 Number of adults 3431 2.858 1.193 1 11 Household net assets (millions of CNY) 3238 0.168 0.470 -0.586 29.961 Non-agricultural activities by the family (dummy) 3428 0.1 0.299 0 1 Number of relatives visited last spring festival 3410 5.781 6.458 0 100 Average household net assets in the village/community (millions of CNY) 3428 0.170 0.224 -0.024 5.313 Average education year of adults in the village/community 3435 5.838 2.49 0.667 14.231 Population in the village/community 3378 4186.825 4663.566 170 51139 Notes: 1. Authors’ calculation based on the family questionnaire of CFPS 2010. 2. Results are weighted using sample weights provided in the data.
36 Table 3. Summary Statistics: by Have a Genealogy Book or Not Variables Genealogy=0 Genealogy=1 Difference (1) (2) (3) (4) (5) Obs Mean Obs Mean Child labor (dummy) 2501 0.091 897 0.046 -0.046*** Work hours per day 2501 0.618 897 0.302 -0.316*** Age 2501 12.494 897 12.433 -0.061 Han nationality (dummy) 2501 0.818 897 0.878 0.060*** Boy (dummy) 2501 0.532 897 0.553 0.021 Rural Hukou (dummy) 2477 0.755 893 0.816 0.061*** Education of father (year) 2439 6.839 881 7.409 0.570*** Education of mother (year) 2451 5.539 872 5.870 0.331 Age of father 2479 40.262 889 40.197 -0.065 Age of mother 2470 38.427 883 38.249 -0.178 Household head is male (dummy) 2501 0.728 895 0.831 0.103*** Number of children 2500 1.778 894 1.933 0.155*** Number of adults 2500 2.852 894 2.864 0.012 Household net assets (millions of CNY) 2355 0.161 857 0.186 0.025 Non-agricultural activities by the family (dummy) 2501 0.101 897 0.098 -0.002 Numbers of relatives visited last spring festival 2488 5.317 892 6.921 1.605*** Average household net assets in the village/community (millions of CNY) 2497 0.172 894 0.156 -0.016** Average education year of adults in the village/community 2501 5.855 897 5.793 -0.062 Log Population in the village/community 2451 7.904 891 7.875 -0.029 Notes: 1. Authors’ calculation based on the family questionnaire of CFPS2010. 2. T-test of difference between children from family with a genealogy book and without is reported in Column (5). 3. Results are weighted using sample weights provided in the data.
37 Table 4. Impacts of Clan Culture on Incidence of Child Labor (1) (2) (3) (4) All Patriarchal and Patrilineal VARIABLES All Girl Boy Genealogy*Boy -0.039* (0.021) Genealogy (dummy) -0.027** -0.006 -0.016 -0.039** (0.012) (0.018) (0.020) (0.015) Age -0.089 -0.088 -0.134 -0.074 (0.059) (0.059) (0.093) (0.076) Age squared 0.003 0.003 0.005 0.003 (0.002) (0.002) (0.004) (0.003) Han nationality (dummy) -0.035* -0.035* -0.066** -0.001 (0.021) (0.021) (0.033) (0.027) Boy (dummy) -0.011 0.000 (0.011) (0.013) Rural Hukou (dummy) -0.006 -0.005 0.002 0.005 (0.019) (0.019) (0.025) (0.026) Age of father -0.006 -0.006 0.053** -0.028 (0.018) (0.018) (0.021) (0.021) Age of father squared 0.000 0.000 -0.001** 0.000 (0.000) (0.000) (0.000) (0.000) Age of mother 0.030** 0.029** 0.001 0.043*** (0.014) (0.014) (0.028) (0.015) Age of mother squared -0.000** -0.000* 0.000 -0.000*** (0.000) (0.000) (0.000) (0.000) Household head is male (dummy) 0.012 0.012 0.020 0.013 (0.014) (0.013) (0.018) (0.020) Number of children 0.003 0.002 0.004 -0.004 (0.010) (0.010) (0.014) (0.013) Number of adults -0.009 -0.009 -0.012 -0.004 (0.006) (0.006) (0.009) (0.007) Household net assets 0.003 0.004 -0.001 0.000 (0.007) (0.007) (0.024) (0.007) Non-agricultural activities by the family (dummy) -0.021 -0.022 0.000 -0.033* (0.015) (0.015) (0.027) (0.018) Number of relatives visited last spring festival -0.001 -0.001 0.002 -0.006*** (0.001) (0.001) (0.002) (0.002) Average household net assets in the village/community -0.034 -0.033 -0.111 0.014 (0.045) (0.045) (0.098) (0.049) Average education year of adults in the village/community -0.014*** -0.014*** -0.002 -0.020*** (0.005) (0.005) (0.009) (0.006) Log population in the village/community 0.025** 0.025** 0.019 0.019 (0.010) (0.010) (0.016) (0.013) Constant -0.021 -0.022 -0.438 0.157 (0.376) (0.376) (0.638) (0.464) Observations 3,023 3,023 1,509 1,514 R-squared 0.251 0.252 0.323 0.321 County FE Y Y Y Y Education of Mother (category) Y Y Y Y Education of Father (category) Y Y Y Y Notes: 1. The dependent variable is children’s labor force participation. 2. Estimates are weighted using sample weights provided in the data. 3. Robust standard errors in parentheses are clustered at household level. 4. *** p<0.01, ** p<0.05, * p<0.10.
44 Table 11. North-South Difference (1) (2) (3) (4) (5) (6) VARIABLES North North&Girl North&Boy South South&Girl South&Boy Genealogy (dummy) 0.007 0.029 -0.028 -0.045** -0.031 -0.045** (0.016) (0.024) (0.019) (0.018) (0.027) (0.022) Observations 1,589 805 784 1,434 704 730 R-squared 0.172 0.280 0.235 0.299 0.367 0.378 Individual char. Y Y Y Y Y Y Household char. Y Y Y Y Y Y Village/Community char. Y Y Y Y Y Y County FE. Y Y Y Y Y Y Notes: 1. The dependent variable is children’s labor force participation. 2. Estimates are weighted using sample weights provided in the data. 3. Robust standard errors in parentheses are clustered at household level. 4. *** p<0.01, ** p<0.05, * p<0.10.
45 Table 12. Mechanism: A Supplement to Formal Institutions (1) (2) (3) (4) (5) (6) VARIABLES Urban Urban&Girl Urban&Boy Rural Rural&Girl Rural&Boy Panel A: by Hukou Status Genealogy (dummy) -0.024 0.007 -0.019 -0.025* -0.014 -0.039** (0.030) (0.040) (0.033) (0.014) (0.021) (0.019) Observations 612 307 305 2,411 1,202 1,209 R-squared 0.457 0.716 0.627 0.271 0.363 0.314 Panel B: by Residence Genealogy (dummy) -0.040 -0.049 -0.035 -0.027** -0.009 -0.044** (0.028) (0.033) (0.025) (0.012) (0.023) (0.020) Observations 715 343 337 3,023 1,136 1,146 R-squared 0.398 0.656 0.516 0.251 0.353 0.315 Notes: 1. The dependent variable is children’s labor force participation. 2. Estimates are weighted using sample weights provided in the data. 3. Robust standard errors in parentheses are clustered at household level. 4. *** p<0.01, ** p<0.05, * p<0.10.
46 Table 13. Mechanism: Impacts of Clan Culture by Household Socioeconomic Status (1) (2) (3) (4) (5) (6) (7) (8) (9) Net asset<1/3 1/3<=Net asset<2/3 Net assets>=2/3 VARIABLES All Girl Boy All Girl Boy All Girl Boy Genealogy (dummy) -0.084*** -0.053 -0.101*** 0.004 -0.008 0.014 -0.003 -0.004 -0.024 (0.021) (0.038) (0.030) (0.026) (0.042) (0.038) (0.024) (0.042) (0.030) Observations 1,004 510 494 1,007 496 511 1,007 502 505 R-squared 0.354 0.515 0.426 0.358 0.485 0.448 0.468 0.624 0.619 Individual char. Y Y Y Y Y Y Y Y Y Household char. Y Y Y Y Y Y Y Y Y Village/Community char. Y Y Y Y Y Y Y Y Y County FE. Y Y Y Y Y Y Y Y Y Notes: 1. The dependent variable is children’s labor force participation. 2. Estimates are weighted using sample weights provided in the data. 3. Robust standard errors in parentheses are clustered at household level. 4. *** p<0.01, ** p<0.05, * p<0.10.
47 Table 14. Mechanism: Risk-sharing (1) (2) (3) VARIABLES All Girl Boy Genealogy (dummy) 0.025 0.034 0.015 (0.024) (0.039) (0.030) Disaster losses at province level 0.004*** 0.004*** 0.005*** (0.001) (0.001) (0.001) Genealogy* Disaster losses at province level -0.003*** -0.003 -0.003** (0.001) (0.002) (0.001) (0.001) (0.001) (0.001) Observations 3,023 1,509 1,514 R-squared 0.111 0.113 0.144 Individual char. Y Y Y Household char. Y Y Y Village/Community char. Y Y Y Provincial char. Y Y Y Region FE Y Y Y Notes: 1. The dependent variable is children’s labor force participation. 2. Estimates are weighted using sample weights provided in the data. 3. Robust standard errors in parentheses are clustered at household level. 4. Provincial characteristics includes log of GDP and log of population in 2009. 5. Six regions: Northeast, Northwest, North China, Southwest, Central South, and East China. 6. *** p<0.01, ** p<0.05, * p<0.10.
48 Table 15. Mechanism: Easing Credit Constraints (1) (2) (3) (4) (5) (6) Father Mother VARIABLES All Girl Boy All Girl Boy Panel A: Getting aids from relatives Genealogy 0.023 -0.033 0.080* -0.020 -0.070 0.001 (0.035) (0.050) (0.048) (0.033) (0.046) (0.045) Observations 2,245 1,106 1,139 2,530 1,264 1,266 R-squared 0.207 0.305 0.266 0.204 0.305 0.265 Panel B: Financial support Genealogy 0.024 -0.042 0.079* -0.054* -0.089** -0.038 (0.035) (0.049) (0.048) (0.030) (0.040) (0.042) Observations 2,221 1,097 1,124 2,497 1,251 1,246 R-squared 0.209 0.314 0.273 0.219 0.317 0.258 Panel C: Aids on other affairs Genealogy -0.034 -0.014 -0.044 0.021 -0.001 0.036 (0.026) (0.038) (0.033) (0.026) (0.039) (0.034) Observations 2,245 1,106 1,139 2,530 1,264 1,266 R-squared 0.158 0.211 0.242 0.162 0.231 0.234 Notes: 1. The dependent variable is whether fathers/mothers receive aids for the following affairs: borrowing money; children’s schooling; seeing a doctor; job searching; children’s job searching. 2. Estimates are weighted using sample weights provided in the data. 3. Robust standard errors in parentheses are clustered at household level. 4.*** p<0.01, ** p<0.05, * p<0.10.
49 Table 16. Mechanism: Social Norms (1) (2) (3) (4) (5) (6) Log educational expenditure (OLS) Study time (Tobit) VARIABLES All Girl Boy All Girl Boy Genealogy (dummy) 0.222** 0.003 0.318** 0.286** 0.188 0.314* (0.103) (0.159) (0.130) (0.143) (0.217) (0.169) Observations 3,007 1,500 1,507 3,023 1,509 1,514 R-squared 0.367 0.437 0.404 Individual char. Y Y Y Y Y Y Household char. Y Y Y Y Y Y Village/Community char. Y Y Y Y Y Y County FE. Y Y Y Y Y Y Log pseudolikelihood -182000000 -83800000 -94900000 Pseudo R-square 0.0797 0.0952 0.0952 Notes: 1. The dependent variable in Columns (1)-(3) is log household educational spending, and in Columns (4)-(6) is hours spending on study. Estimates are weighted using sample weights provided in the data. 2. Robust standard errors in parentheses are clustered at household level. 3. *** p<0.01, ** p<0.05, * p<0.10.
TheUNU‐MERITWORKINGPaperSeries 2022-01 Structuraltransformationsandcumulativecausationtowardsanevolutionary micro‐foundationoftheKaldoriangrowthmodelbyAndréLorentz,TommasoCiarli, MariaSavonaandMarcoValente 2022-02 Estimationofaproductionfunctionwithdomesticandforeigncapitalstockby ThomasZiesemer 2022-03 Automationandrelatedtechnologies:Amappingofthenewknowledgebaseby EnricoSantarelli,JacopoStaccioliandMarcoVivarelli 2022-04 Theold‐agepensionhouseholdreplacementrateinBelgiumbyAlessioJ.G.Brown andAnne‐LoreFraikin 2022-05 GlobalisationincreasedtrustinnorthernandwesternEuropebetween2002and 2018byLoesjeVerhoevenandJoRitzen 2022-06 GlobalisationandfinancialisationintheNetherlands,1995–2020byJoanMuysken andHuubMeijers 2022-07 Importpenetrationandmanufacturingemployment:EvidencefromAfricaby SolomonOwusu,GideonNdubuisiandEmmanuelB.Mensah 2022-08 AdvanceddigitaltechnologiesandindustrialresilienceduringtheCOVID‐19 pandemic:Afirm‐levelperspectivebyElisaCalzaAlejandroLavopaandLigiaZagato 2022-09 Thereckoningofsexualviolenceandcorruption:Agenderedstudyofsextortionin migrationtoSouthAfricabyAshleighBickerCaarten,LoesvanHeugtenandOrtrun Merkle 2022-10 TheproductiveroleofsocialpolicybyOmarRodríguezTorres 2022-11 SomenewviewsonproductspaceandrelateddiversificationbyÖnderNomaler andBartVerspagen 2022-12 ThemultidimensionalimpactsoftheConditionalCashTransferprogramJuntosin PerubyRicardoMorelandLizGirón 2022-13 Semi‐endogenousgrowthinanon‐WalrasianDSEMforBrazil:Estimationand simulationofchangesinforeignincome,humancapital,R&D,andtermsoftrade byThomasH.W.Ziesemer 2022-14 Routine‐biasedtechnologicalchangeandemployeeoutcomesaftermasslayoffs: EvidencefromBrazilbyAntonioMartins‐Neto,XavierCireraandAlexCoad 2022-15 ThecanonicalcorrelationcomplexitymethodbyÖnderNomaler&BartVerspagen 2022-16 CanonicalcorrelationcomplexityofEuropeanregionsbyÖnderNomaler&Bart Verspagen 2022-17 QuantilereturnandvolatilityconnectednessamongNon‐FungibleTokens(NFTs) and(un)conventionalassetsbyChristianUrom,GideonNdubuisiandKhaled Guesmi 2022-18 HowdoFirmsInnovateinLatinAmerica?IdentificationofInnovationStrategiesand TheirMainAdoptionDeterminantsbyFernandoVargas 2022-19 Remittancedependence,supportfortaxationandqualityofpublicservicesinAfrica byMatyKonteandGideonNdubuisi 2022-20 HarmonizedLatinAmericaninnovationSurveysDatabase(LAIS):Firm‐level microdataforthestudyofinnovationbyFernandoVargas,CharlotteGuillard, MónicaSalazarandGustavoA.Crespi 2022-21 Automationexposureandimplicationsinadvancedanddevelopingcountriesacross gender,age,andskillsbyHubertNii‐Aponsah
2022-22 SextortioninaccesstoWASHservicesinselectedregionsofBangladeshbyOrtrun Merkle,UmrbekAllakulovandDeboraGonzalez 2022-23 Complexityresearchineconomics:past,presentandfuturebyÖnderNomaler& BartVerspagen 2022-24 Technologyadoption,innovationpolicyandcatching‐upbyJuanR.Perillaand ThomasH.W.Ziesemer 2022-25 ExogenousshocksandproactiveresilienceintheEU:ThecaseoftheRecoveryand ResilienceFacilitybyAnthonyBartzokas,RenatoGiaconandCorradoMacchiarelli 2022-26 Achanceforoptimism:Engineeringthebreakawayfromthedownwardspiralin trustandsocialcohesionorkeepingthefishfromdisappearingbyJoRitzenand EleonoraNillesen 2022-27 Dynamicdependencebetweencleaninvestmentsandeconomicpolicyuncertainty byChristianUrom,HelaMzoughi,GideonNdubuisiandKhaledGuesmi 2022-28 PeernetworksandmalleabilityofeducationalaspirationsbyMichelleGonzalez Amador,RobinCowanandEleonoraNillesen 2022-29 LinkingtheBOPCgrowthmodelwithforeigndebtdynamicstothegoodsandlabour markets:ABOP‐IXSM‐OkunmodelbyThomasH.W.Ziesemer 2022-30 Countries’researchprioritiesinrelationtotheSustainableDevelopmentGoalsby HugoConfraria,TommasoCiarliandEdNoyons 2022-31 MultitaskingbyAnzelikaZaiceva‐Razzolini 2022-32 Informalinstitutionmeetschilddevelopment:ClancultureandchildlaborinChina byCanTangandZhongZhao