1 Introduction

The official from the Colombian Institute of Family Welfare asked the agent to stop at the corner of Carrera 51 and Calle 85, to make the routine inspection of the mobile stands that are usually installed in the area. Every week they find some cases to report to the Childhood and Adolescence Police of the Barranquilla City Hall. On this occasion, a child of about 9 years old is in charge of a humble stand with paper handkerchiefs, mineral water, soursop juice, and “lulo” candies.

Jairo was a little scared when the police started questioning him, so he answered all his questions frankly, one by one. That morning he woke up early in the house in Pinar del Rio. His mother’s cousin picked him up and brought him in a mule cart to the north of the city. He says he lives with his mother and stepfather. He explains that the situation changed drastically when his father abandoned them a few years ago. “My mother doesn’t pay attention to me and sends me out on the streets to sell with her cousin. On a normal day, I stay at the stand and sell until late afternoon, when they come to pick me up on their way home. Many times I have been left alone on the street for several days. The man my mother lives with hits me and I prefer to stay outside until things calm down a bit”.

Many children work on the streets under the supervision of an adult relative, to whom they can turn if they have a problem. These children give most of their income to their families. However, several of Jairo’s neighbors have already moved onto the streets on their own, attracted by the possibility of earning an income of their own, or seduced by the drugs and entertainment opportunities the city has to offer. To earn a living, they sell mobile phone cards, clean car windows at traffic lights, sell drinks and sweets, or play the role of informal shoe polishers. Pinar del Río is a settlement formed by people displaced by political violence, where mostly poor families live. For many children, there is no great difference between sleeping on the street and sleeping on the floor of their house, in a house without electricity or water.

In this chapter, we look at the causes and consequences of child labor. As Jairo’s case reveals, children who work on the street are highly likely to come from dysfunctional families, have experienced geographic displacement, or live in poverty. Even so, this is a very heterogeneous group: some live permanently on the street; others live at home, even if they work on the street; and many are in a constant flow between the street and the home. Among other factors, domestic violence, low family income, father’s alcoholism, neglect, and abuse are among the antecedents of this form of child labor.

We then describe the most common causes of different types of child labor and examine some of the most relevant consequences, paying particular attention to the educational impact and the repercussions on employability during adult life. In Table 2.1 we have summarized the set of factors that we are going to review in the following pages.

Table 2.1 Background and consequent factors of child labor
Full size table

2 Social, Economic, and Educational Antecedent Factors to Child Labor

Child labor is considered both a cause and a consequence of the lack of economic development (Edmonds, 2016). On the one hand, the rate of child labor is directly related to the gross domestic product. On the other hand, there is evidence that a country’s economic development reduces the rate of child labor. At the macroeconomic level, this process runs parallel to the reduction of the relative weight of the primary sector in the economyFootnote 1 (and is also partly associated with technological development).

At the microeconomic level, increased family income improves living standards and, in the long term, reduces child labor. The change is more significant in the poorest households (Edmonds, 2005). In the short term, however, there may be a temporary upturn, when households involve their children in work activities to take advantage of new opportunities (Kambhampati & Rajan, 2006). It is therefore advisable to adopt a perspective that transcends economic cycles.

In this context, a widespread type of study consists of an econometric analysis of the determinants of child labor. In the first part of this chapter, we review in some detail some of the most relevant background factors. In the second part, although it has had comparatively less weight in research on the topic, we summarize some of the existing evidence on the impact on health, educational experience, and future work.

2.1 Poverty and Child Labor

Family poverty is the first direct antecedent that pushes children to work. It is usually assumed that adults only send their children to work when their own income falls to very low levels (Basu & Van, 1998; Fan, 2011). Consequently, household poverty is positively related to the likelihood of children working and negatively related to schooling (Amin et al., 2004; Ray, 2000; Triningsih & Ichihashi, 2010; Verner & Blunch, 1999). Keeping children away from work is a luxury that poorer families cannot afford.

Nonetheless, the evidence for this is not always consistent and the size of the effect may be small (Fan, 2011; Wahba, 2000). This has led to some efforts to demonstrate the “transmission of poverty”, with the understanding that child labor contributes to the intergenerational perpetuation of poverty. Also, there may be other family, educational, or social variables that modulate the relationship between household income and child labor (Holgado et al., 2016).

Box 2.1: The Axiom of Luxury, the Substitution of Adult Labor and the Paradox of Wealth

Research has connected household income with the likelihood that children will work in at least two different ways: (1) poverty as a cause of child labor: low household income increases the likelihood that children will work, (2) transmission of poverty: children are more likely to work if their parents worked as children. Basu and Van (1998) theorized about this, assuming that parents do not take advantage of it for selfish interests but are forced to have their children work for survival needs. On paper, the prevalence of child labor could have a negative impact on the working conditions of adults. Based on these previous approaches, they introduced two operating principles:

  • The axiom of luxury: “a family will send children into the labor market only if the family income from non-child labor sources falls very low”.

  • The axiom of substitution: “from a company’s point of view, child labor and adult labor are substitutes”.

However, in the rural contexts of developing countries, it has been found that sometimes a better relative position of the family is associated with the fact that their children work. This is particularly the case in the agricultural sector where a subsistence economy predominates, so an alternative operating principle can be introduced (Bhalotra & Heady, 2003):

  • The paradox of wealth: “children from land-rich households are often more likely to work than those from land-poor households”.

This has led to the idea that the substitutability of adult work for child labor is more important as an explanatory factor than the absolute income of the parents.

Based on: Fan, C. S. (2011). The luxury axiom, the wealth paradox, and child labor. Journal of Economic Development, 36(3), 25–45.

2.2 Family Background

The educational level of the parents and the fact that the parents worked as children are also predictive factors to take into consideration. A study in Egypt found that having a parent who worked as a child doubled the likelihood of a child entering the labor market. Moreover, the influence was much greater when it was the mother who had worked during her childhood (Wahba, 2000). Consistent with this, data from India showed that the higher the mother’s educational level, the less likely her children are to work (Mukherjee & Das, 2008). Being educated seems to influence parents to place more importance on children’s schooling.

The intergenerational transmission of the educational level can be reflected in a limitation of the opportunities available in the labor market. It not only generically reduces social mobility, for women it is one of the factors influencing the gender gap (Van Putten et al., 2008).

2.3 The Characteristics of the Families

Econometric models have controlled for the effect of household income by incorporating other household characteristics that may have an impact, such as the number of siblings, the number of people in the household, and the order of birth. The size of the family group facilitates the specialization of its members so that some children work while others study (Patrinos & Psacharopoulos, 1997). In particular, it is more common for first-born children to be involved in work activities and to contribute to the family income so that younger siblings can devote themselves to study (Emerson & Souza, 2008).

On the other hand, both exposures to an armed conflict and forced displacement of the family are risk factors for school dropouts and early incorporation into the labor market (Holgado et al., 2016; Rodriguez & Sanchez, 2012). Similarly, rural-urban migration results in the resettlement of large segments of the population on the outskirts of urban centers, where they encounter problems of access to basic resources and live in conditions of poverty and social exclusion (Plummer et al., 2007; Siddiqi & Patrinos, 1996). The risk is greater for unaccompanied minors.

In contrast, the international migration of parents can have a protective effect on the child labor of children left behind (Antman, 2012; Mansuri, 2006), especially in the case of mothers (Acosta, 2011a). Remittances sent by immigrants make it possible to invest in children’s human capital (Acosta, 2011b; Alcaraz et al., 2012).

2.4 The Quality of the Education System

Shortcomings in the education system may be a precursor to child labor. This refers to the quality of teaching, teacher training, accessibility for high-risk groups, institutional capacity to guarantee compulsory education, teaching conditions, and, where appropriate, the overcrowding of classes or the lack of adequate infrastructure. A fairly informative fact in this regard is usually the salary of primary and secondary school teachers, as well as their working conditions. When the education system functions poorly, a higher proportion of working children is often found (Canagarajah & Coulombe, 1999; Canagarajah & Nielsen, 1999). In contrast, programs to improve the quality of education not only increase school attendance but indirectly discourage the involvement of children in work activities, especially at the secondary level (Rossi & Rosati, 2007).

2.5 The Culture of Work and the Attitudes of the Population

Finally, several cultural factors should be mentioned among the determinants, such as prevailing population values, parental beliefs, and social perception of child labor. In many families, girls are still expected to engage in domestic activities, just as boys are expected to help on the family farm in the countryside. Parents have different degrees of concern about the safety of their children (Runyan et al., 2009); and they also differ in their expectations about the role that boys and girls should play (Buchmann, 2000), or in the impact that work may have on their learning for life (Bahar, 2014). The degree of society’s tolerance for children’s work may also vary considerably between countries (Bahar, 2014; Maya Jariego, 2017).

Box 2.2: Two Profiles of Child Labor

Child labor takes different forms depending on the social and community context in which it occurs. In a survey of 3259 families in the Colombian Caribbean, two profiles were identified based on background factors for child labor involvement. First, in families living in urban areas, child labor was significantly related to family income and the socio-labor characteristics of the mothers. Specifically, children were more likely to work when their mothers were unemployed, they had no education or they had worked as children. In any case, family income was the best predictor of child labor. Therefore, the background of this profile corresponds to some of the key factors identified in the previous literature on the subject.

However, in families living in a rural or peri-urban environment (on the outskirts of large cities) the most relevant predictor was having experienced forced displacement in the last 5 years. Migration to the city is a situation of risk of child labor, which generates special needs. This risk profile can be enhanced when symptoms of post-traumatic stress occur among those who have been victims of political violence. Colombia is one of the countries with the largest number of internally displaced persons in the world, so child labor is necessarily connected to this phenomenon in a significant part of the population.

Based on Holgado, D., Maya-Jariego, I., Palacio, J. & Oviedo-Trespalacios, O. (2016). Two profiles of child labor in the Colombian Caribbean Coast: relocated children to suburban areas compared to the key role of social and labor characteristics of mothers in urban settings. In Tonón, G (Ed.). Indicators of Quality of Life in Latin America, pp. 251–273. Social Indicators Research Series, Vol. 62. New York: Springer.

[Colombia]

3 The Impact on Health, Educational Experience, and Future Employment

The existence of a large pool of child workers can have a negative impact on economic development, either by indirectly reducing the wages of unskilled labor or by discouraging the adoption of knowledge- or skills-intensive technology (Edmonds, 2016). By definition, child labor generates an impact on the human capital of the younger generation. In turn, the impact on children’s training, development, and capabilities is reflected in the long term in a reduced take-up of opportunities for economic growth.

In this section, we will pay special attention to the impact on the educational field, where most evidence has been accumulated. Complementarily, we gather the available information on the consequences on health, welfare, and employability.

3.1 Risk, Health, and Mortality

Child labor can be associated with both short-term and long-term health problems. Much of the research has focused on assessing the impact of performing hazardous work, along with the combined effect of dropping out of studies. The risks are highly variable depending on the activity, ranging from exposure to pesticides in agricultural work, or inhalation of toxic substances during shoemaking, to broken bones during fruit picking on trees, or injuries caused by machinery.Footnote 2 Although there are no reliable statistics and many incidents are simply not reported, occupational accidents and child mortality rates appear to be at least comparable to those of adults. In turn, delaying entry into the workforce reduces the likelihood of the early onset of back pain, arthritis, or reduced physical strength (Lee & Orazem, 2010). With data aggregated by country, evidence has been found of the relationship between child labor and adolescent mortality, the level of nutrition of the population, and the presence of infectious diseases (Roggero et al., 2007).

Two recent systematic reviews link child labor with growth problems, malnutrition, a higher incidence of infectious diseases, a high prevalence of injuries, behavior problems, and a lower coping capacity (Batomen Kuimi et al., 2018; Ibrahim et al., 2019).

3.2 Psychological Well-Being and Quality of Life

There are comparatively few studies analyzing the consequences of child labor on children’s mental health. Some epidemiological studies have documented a higher relative prevalence of mood problems and separation anxiety (Benjet, 2010; Doocy et al., 2007; Fekadu et al., 2006), and for girls a higher likelihood of family violence (Catani et al., 2009). Engaging in risky activities and long hours can generate feelings of frustration, so many children become withdrawn and uncommunicative (Uddin et al., 2009). In the case of child soldiers, ex-combatants experience symptoms of post-traumatic stress, are more vulnerable to drug abuse, and may have more difficulties with social interaction and community integration (Schauer & Elbert, 2010).

3.3 Academic Performance and School Dropout

In the educational field, the two most contrasting results are school dropouts and deteriorating academic performance. In the first place, working time seems to displace the time dedicated to studies (Fernandez & Abocejo, 2014) and can reduce the period of schooling by up to 2 years (Psacharopoulos, 1997). Dropout rates may increase when local labor market opportunities improve (Duryea & Arends-Kuenning, 2003; Kruger, 2007), and are generally higher in rural contexts (Canagarajah & Coulombe, 1999). In urban areas, they are concentrated in disadvantaged neighborhoods with higher levels of social risk (Crane, 1991).

Second, working children perform worse in language and mathematics (Gunnarsson et al., 2006), and are more likely to repeat grades (Psacharopoulos, 1997). The negative impact in terms of learning may be due to fatigue and “the diversion of their interests away from academic concerns” (Heady, 2003). In addition, it is directly related to the number of hours dedicated to work and the performance of work activities in a schedule that coincides with that of the school (Holgado et al., 2016). Jobs that are more intense or require more dedication may also have a more pronounced impact on academic performance.

Results may also depend in part on family characteristics. In large families, there is a greater likelihood of dropouts or poorer educational outcomes. There seems to be a kind of “specialization” in the household so that some children are dedicated to studying and others are not (Patrinos & Psacharopoulos, 1995). A study in Peru found that the number of siblings was associated with the probability of repeating a grade (Patrinos & Psacharopoulos, 1997).

In addition, when making decisions about their children’s education, families often also consider educational facilities, the number of teachers, the percentage of female teachers, and the distance to school (Huisman & Smits, 2009). When there are more teachers in the district of residence, children are more likely to go to school. In the specific case of girls, parents also positively value the availability of female teachers. Finally, in many cases, geographical distance and lack of adequate transport infrastructure remain a barrier to access, especially for low-income families.

3.4 Employability, Job Opportunities, and Working Conditions

It is often suggested in this area that having worked during childhood may contribute to unemployment and low salaries among adults (Bequele & Boyden, 1988), although there is little empirical development to demonstrate this. In the short term, the work involvement of children increases household income and the probability of survival, but in the long term, it can contribute to the perpetuation of poverty to the extent that it reduces “human capital accumulation” (Galli, 2001). Early engagement in unskilled work (whether in agriculture or industry) reduces opportunities for skills development and later translates into lower-paid jobs during adulthood.

In a study with longitudinal data in Tanzania, working during childhood was found to be a significant predictor of staying in the agricultural sector 10 years later (Beegle et al., 2008). More specifically, when children worked between 10 and 20 h/week, they were more likely to be in vulnerable jobs as adults. Besides, working on the family farm had the most negative effects on girls (Burrone & Giannelli, 2019).

Recently, the “Global Report on Child Labour 2015” (ILO, 2015) has also attempted to document the connection between working during childhood and employment outcomes during youth. An extensive survey examining the transition from school to work found that individuals who had worked as children showed greater youth vulnerability, along with more difficulty in finding good jobs. Also, it revealed that adolescence is a high-risk profile, with a high prevalence of “hazardous work” between the ages of 15 and 17.

These types of studies that follow the trajectory of working children through the life cycle are likely to become more frequent in the coming years. Not only would they serve to analyze the long-term impact, but they are also useful for assessing the equation between protective factors and cumulative risk factors.

Box 2.3: At the Street School

Environmental psychology has studied individual differences in the representation of space, to then determine how it influences movements in complex urban environments. One of the most frequent case studies consists of analyzing the “cognitive maps” of taxi drivers since they accumulate experience of movement within the city that is clearly above that of the average citizen. It has been found that taxi drivers can estimate shorter travel distances than the general public, among other things because they know more shortcuts. Therefore, it seems that their professional experience allows them to develop enriched representations of the street map of the locality in which they work. It would be interesting to see whether the widespread use of GPS has in any way undermined their ability to navigate efficiently in the city.

What about working children? Do they learn anything in the “street school”? Some research seems to indicate this. A study with children in the Northeast of Brazil showed that those who sold sweets on the street acquired levels of mathematical understanding that in some dimensions were above the comparison group. Specifically, the vendors developed adequate strategies to solve arithmetic problems and were better able to handle large numerical values. In this case, cognitive development seems to be connected to participation in everyday buying and selling practices.

Adult traders often jokingly say that they have “graduated from the psychology of the street” because they are in constant contact with people of all kinds who test their communication skills. This is another area in which it would be of interest to explore the skills developed by children working on the street.

Based on Saxe, G. B. (1988). The mathematics of child street vendors. Child Development, 1415–1425.

[Brazil]

4 The Cycle of Poverty

Research on the causes and consequences of child labor allows us to draw a hypothetical cycle of poverty, as we have represented in Fig. 2.1. This sequence is based on some of the evidence we have found in our previous review of the literature. Families with low income are more likely to send their children to work. This can interfere with the continuation of compulsory education and therefore has a negative impact on the type of skills that allow upward labor mobility. As a result, children are highly likely to reproduce as adults the same kind of subsistence activities that their parents performed. This closes the intergenerational cycle that explains the persistence of poverty in certain segments of the population.

Fig. 2.1
figure 1

The hypothetical cycle of poverty. Source: own elaboration

Full size image

This is a cycle that is especially documented in children who carry out non-salaried activities in the family unit, collaborating in tasks generally in the primary sector. Especially in the case of subsistence agriculture. The “vicious circle of poverty” is also widely documented in development economics. It has been shown that lack of income prevents the generation of savings and thus the accumulation of capital needed to increase income (Bauer, 1965). The result is that societies with higher levels of income inequality tend to have lower levels of intergenerational income mobility as a consequence (Sakamoto et al., 2014).

At the micro-level, something similar seems to be happening. To the extent that children from low-income households have to spend their time helping the family, they cannot free up the time needed to invest in the human capital that would improve their positioning in the labor market, and that would potentially allow them to reverse their situation. From a psychological point of view, both the accumulation of risk factors and continued exposure to stressors can result in some deterioration of cognitive abilities (Sakamoto et al., 2014). It also often results in the development of fatalistic attitudes and the generalization of feelings of learned helplessness.

5 Conclusion

Child labor has a higher incidence in low-income families, so it seems related to situations of poverty. However, this is a complex relationship. On the one hand, there does not seem to be a simple linear correlation between family income and the probability of children working, but rather it is modulated by family characteristics as well as other social and cultural factors. For example, parents’ educational expectations, along with their own history of child labor, carry significant weight. On the other hand, econometric models have been specifically based on economic survival activities, rather than on forced or dangerous work, so it would be necessary to check to what extent they can be generalized to this type of situation as well.

Secondly, we have described some of the mechanisms that seem to underpin the intergenerational perpetuation of poverty through the impact of child labor on education. In the long term, one can follow the implications of educational interruption on adult salaries, incomes, and poverty (Ilahi et al., 2000, 2009). If in Chap. 1 we presented prevention as a strategy to avoid the aggravation among the various forms of child labor, in this chapter we discovered its potential to avoid the worsening or intergenerational permanence of child labor in certain segments of the population. As we shall see below, actions to ensure universal primary and secondary education are directly related to poverty reduction.

The consequences of child labor seem to be grouped around two types of effects: some related to health and well-being, while others are related to education and employability. Research on health risks has paid more attention to those considered “hazardous work”, while assessment of the impact on education has been comparatively more frequent in primary sector livelihood activities. However, there is a need to systematically unravel what effects are associated with each form of child labor. The comparative study of modalities and effects can contribute to a better understanding of the phenomenon.

Finally, we have identified at least two topics for further research in the future. First, although negative consequences are part of the very definition of child labor, their empirical documentation has been relatively scarce, so it would be interesting to delve deeper into this aspect in the future. Second, as research in other areas has shown, protective factors contribute to the resilience of a significant proportion of those individuals who experience an accumulation of risks in their childhood. Consequently, it would be of interest to explore which elements specifically contribute to the resilience of working children.