Low- and middle-income countries (LMICs) often face significant household shocks due to their vulnerability to natural disasters, economic fluctuations, and limited access to social safety nets [1-3]. Child labour has been associated with various shocks, such as climate-related disasters like floods and droughts, changes in food prices, natural disasters, and parents' health issues [4,5]. Additionally, most households in these countries depend on agriculture as their primary source of income, making them particularly susceptible to household shocks [6]. This reliance on agriculture has led to increased income instability. Furthermore, credit and insurance markets in low-income countries (LICs) often need to be developed, leaving households with limited coping mechanisms for unexpected shocks [7].In response, households in LMICs often employ various coping strategies to maintain their living standards, including increased reliance on child labour [8,9] defined household shocks as unanticipated and adverse events resulting in loss of income, reduced consumption, or loss of assets for a household. These shocks are linked to child labour in several ways. Jensen [9] and Skoufias [10] suggest that the primary impact of shocks on households is a decrease in income, compelling families to seek alternative sources of revenue, including the labour contributions of their children. The need to involve children in income-generating activities increases as households experience shocks and face higher expenses for essentials [11]. However, household shocks can also disrupt the support networks within communities and social networks. Skoufias [10] explains that this situation can leave households with fewer options to cope with income fluctuations. In these countries, the absence of institutional credit or insurance markets makes it challenging to address the issue of child labour [12]. Moreover, there is no protection against unforeseen death, and families often resort to informal coping strategies, such as child labour, to deal with shocks. The effects of shock-induced child labour are wide-ranging and significant. Child labour often involves physically demanding tasks, hazardous conditions, and exposure to harmful substances, leading to various health issues [13]. Short- and long-term health consequences include injury, respiratory problems, and malnutrition. These health issues can result in chronic illnesses or disabilities, limiting the child's prospects for a healthy and fulfilling life [14]. Further, child labour hinders children's access to education, as they typically lack the time or energy to attend school. Evidence indicates that school attendance declines when child labour hours increase [15, David [16]. This can lead to poor reading and numeracy skills, high dropout rates, and low educational achievements [8,17]. Such conditions adversely affect the children's future potential and earnings. The prevalence of child labour substantially affects a country's rate of progress. An undereducated workforce and poor health due to child labour tend to be less productive, potentially hindering a nation's economic growth [18]. Moreover, child labour facilitates the intergenerational transmission of poverty because of its effects on children's education and skill levels, limiting their opportunities for upward mobility. To safeguard the well-being of children and ensure the nation's future development, the country must address this problem. This paper examines the effects of household shocks such as climatic shocks (including floods and drought), parental death shocks, and rising food prices on child labour in Tanzania. Droughts and floods are climate shocks that can significantly disrupt agricultural production, household income, and food security. Research in Tanzania has shown that these climate shocks contribute to increased child labour, as families rely on their children to help with agricultural work or find alternative income sources [19]. In addition to the delayed effects observed in other developing countries, droughts and floods have increased the likelihood that children will work. For example, Delap [20] found that children in Bangladesh were more likely to engage in labour activities because of the negative impacts of floods on household income and food security. Likewise, Woldehanna [21] demonstrated that droughts and floods in Ethiopia increase the probability of children working, accompanied by a decline in school attendance. Food price shocks significantly affect child labour as they strain household budgets and push families to seek alternative coping mechanisms, including child labour. Numerous studies from various countries have demonstrated a connection between food price shocks and child labour. D'Souza and Tandon's [22] research in India revealed that high food prices significantly increased the likelihood of children working. This relationship is explained by the need for families to earn more money to afford the rising food costs and maintain their consumption levels. In this context, children become valuable labour resources, helping supplement household income by working in the family business or taking on paid employment outside the home. Similar findings were reported in Pakistan by Bhalotra and Heady [4], who discovered that changes in food prices were associated with an increase in child labour. Their study suggests that children from lower-income households are more likely to feel the effects of food price shocks, as their families have fewer coping mechanisms. Dumas [23] investigated the link between fluctuations in food prices and child labour in Burkina Faso. According to the study, rising food prices were strongly correlated with child labour, particularly among girls. The report also highlights the need for social protection programs to address households' vulnerability to fluctuations in food prices. On the other hand, parental death also significantly impacts child labour. When a parent passes away, the family's resources and income may be substantially reduced, compelling children to enter the workforce and compensate for the loss of income. Numerous studies from various countries have consistently demonstrated a strong relationship between increasing child labour and parental death. Webbink [24] analyzed the effects of parental mortality on children's schooling and employment activities in Rwanda. They discovered that the loss of a parent significantly increased child labour, particularly for orphans who had to work to support their families. Several studies have extensively documented the effects of household shocks on child labour. Beegle [8] assessed the impact of crop shocks on child labour in Tanzania and found a positive correlation between the two variables. Bandara and Lavie-Rouse [19] linked agricultural shocks to child labour, noting an increase in child labour hours, particularly for boys. Huang and Acheampong [12] observed that health shocks affecting men had a more significant impact on children's labour in Nigeria than those affecting women, highlighting the importance of considering gendered aspects of household shocks. Webbink [24] found that children in Rwanda who experienced the loss of a parent were more likely to engage in labour activities and reduce their time spent in school.Similarly, Abou (2014) reported that Nigerian children who lost their parents were more likely to be involved in labour activities and less likely to attend school. However, most of these studies have focused on a single measure of shocks, such as Abou (2014) and Webbink [24] on death shocks, Huang and Acheampong [12] on parental health shocks, and Beegle [8] on crop shocks as a proxy for agricultural shocks. Others, such as Bandara and Lavie-Rouse [19], consider two measures of shocks. While studies in Tanzania, such as Beegle et al. [8] and Bandara and Lavie-Rouse [19], have employed longitudinal datasets, these datasets still need to be updated. This study builds on the literature by extending the work of Bandara and Lavie-Rouse [19] in Tanzania, focusing on three household shocks instead of the two previously studied. Examining multiple shocks allows a more comprehensive understanding of how they influence child labour. In addition, this study uses a more recent panel dataset than Bandara and Lavie-Rouse [19], reflecting current trends and recent economic conditions. This up-to-date information enables policymakers to make informed decisions based on their current situation. This study also fills the gaps in the literature by using reported shocks by households during the survey. Unlike Bandara and Lavie-Rouse [19] and Beegle et al. [8], Tanzania used proxy shock variables as income shocks proxied by crop loss and agricultural shocks. Utilizing reported shocks from households presents multiple advantages over relying on proxy shocks. Reported shocks enable direct measurement of household experiences, providing a more accurate evaluation of their exposure to shocks. They deliver comprehensive coverage of various events that impact households, including health, economic, social, and climatic shocks. This allows for a more holistic understanding of the factors driving child labour. Moreover, reported shocks offer context-specific information, capturing subtleties that proxy measures might overlook. They also uncover heterogeneous effects across households, facilitating a nuanced examination of the relationship between shocks and child labour. Lastly, reported shocks tend to be more timely and accurately represent the current state of household vulnerability than proxy measures [9, 10,25]. The rest of the paper is organized as follows: Section 2 shows the child labour situation in Tanzania, Section 3 shows the methods and data, Section 4 indicates the findings, Section 5 presents the discussion, and Section 6 concludes the paper and provides policy recommendations.

Child labour remains a pressing issue in Tanzania despite the country's efforts to address the problem through legal frameworks and implementation policies. According to the Law of the Child Act of 2002, a child is defined as a person under 18, in line with the Convention of Human Rights No. 138. The International Labour Organization (ILO) defines child labour as allowing countries with underdeveloped economies and educational systems to initially set the legal minimum age at 14 years (12 years for light employment). This study adheres to the ILO's definition of child labour, in which the Employment and Labour Relations Act No. 6 prohibits the employment of children under 14. To support this, the government introduced policies such as the National Plan on the Elimination of Child Labour in 2009, aiming to reduce the incidence of child labour across rural and urban areas and all economic sectors. Furthermore, the Child Development Policy of 2008 implemented the ILO Convention, targeting eliminating the worst forms of child labour. Despite these efforts, child labour remains prevalent in Tanzania, mainly in rural areas. According to the Tanzanian National Population Census Survey data of 2022 statistics, approximately 4.2 million children aged 5-17 are engaged in various child labour activities, equivalent to 28.8% of children in this age group (NBS, 2022). A higher proportion of child labour was observed in rural areas (35.6%) than in urban areas (18%). Approximately 67.1% of child labourers aged 5-17 are involved in housekeeping and economic activities such as agriculture and mining (URT, 2021). Child labourers in Tanzania work approximately 25 hours per week across all age groups and activities (NBS, 2022), limiting their schooling time. Despite implementing legal frameworks and policies, millions of children continue to engage in various labour activities, hindering their education and overall development. Concerted efforts by the government, NGOs, and the international community are needed to tackle this pressing issue and safeguard Tanzanian children's futures.

Child Labour Theories

Numerous theoretical frameworks have been used to study child labour in great detail, illuminating the complexity of this complicated societal issue. The Economic Theory, which contends that families are compelled by poverty and economic hardship to send their children to work in order to supplement the family income, is one of the most widely accepted hypotheses explaining child labour. This viewpoint, which is frequently credited to academics like Edmonds [26] and Basu [27,28], highlights the need for families to economically force their children to participate in labour-intensive activities, particularly in poor nations where financial limitations are very tight.

Ennew and Swart-Kruger [29] introduced the Cultural-Cognitive Theory, which explores the cultural norms and beliefs that support child labour. Because of ingrained customs and social norms, there are some civilizations where it is often accepted for children to begin earning money at a young age. This idea emphasizes how child labour is maintained by deeply rooted cultural practises, which turn it into a norm. However, Bhagwati [30] presented the Globalisation Theory, which looks at how globalization and demands from foreign markets can encourage child labour. According to the hypothesis, businesses in a global marketplace that is competitive look for low-cost labour, frequently abusing underage labourers in developing nations with lax labour laws. This hypothesis focuses on the demand-side elements that contribute to the continuation of child labour by creating a market for it as a result of globalization. The opportunity cost of schooling for children involved in labour is examined by the Human Capital Theory, which was promoted by Becker in [31]. This hypothesis holds that children are pulled out of school because their labour is valued higher than knowledge. This viewpoint, which sees education as an investment that can eventually raise families out of poverty, emphasizes the importance that education plays in ending the cycle of child labour. On the other hand, Bandura [32] explored the Social Learning Theory, which explores the impact of social settings on child labour. Growing up in households or places where labour is accepted as normal can cause children to internalize these attitudes and so continue the cycle. This theory emphasizes how crucial social interventions and awareness campaigns are to changing public perceptions of and expectations around child labour.

Data

This study utilized the National Panel Survey Datasets for 2014–15 and 2019–20. The National Bureau of Statistics in Tanzania collects the NPS, which collaborated with the World Bank and covers both the Tanzanian Mainland and Zanzibar. It is part of the Standard Living Measurement Survey. The fourth and fifth waves employed in this study occurred in 2014–15 and 2019–20, respectively, marking the expanded NPS. In addition to child labour and household data, it gathers information on various socioeconomic characteristics of households, such as age, gender, geography, and marital status. It also details farming characteristics and different work patterns for individuals above five years of age. These data were used in the current paper to examine the relationship between shocks and child labour in Tanzania. The unit of analysis for this study consisted of children aged 7-13 years attending primary school in Tanzania. Students in Tanzania are expected to enter primary school at seven and graduate at 13, provided they are enrolled on time. The legal minimum working age in the country, as established by the International Labour Organization, is 14. Therefore, children below this age range who engage in activities that impede their development are involved in child labour. The study merged the two datasets of the NPS by tracking all children aged 7-13 years at the time of the interviews across both years. A balanced panel of 588 participants was created, monitoring individuals aged 7 to 13 during the 2014–15 and 2019–20 waves. This approach enabled the examination of child labour trends and the impact of household shocks on these children over time.

Outcome Variable

The main outcome variable in this study was child labour, as measured using child hours. Child labour is defined as the total hours spent by a child aged 7-13 years on economic activities for wages, household-run businesses, farming, and unpaid household tasks during the previous week. Measurements of child labour were adopted from previous studies [19,33].

Explanatory Variables

Household shocks were the main explanatory variable in this study. The three household shocks severely affecting the household's income and welfare include weather shocks, the death of a family member, and food price rise shocks. Household shock is measured as a dummy variable, indicating one if the household member experienced shocks two years before the survey date and 0 if otherwise. The selection of the shocks was due to the frequencies (thus, the shock which occurs more frequently was selected).In addition, such shock must severely affect the household’s income and reduce their welfare.

Control Variables

This study incorporated several control variables, including household size, children's age, household head's age, marital status, children's sex, household head's sex, location, and school attendance. Both child and household head ages were measured in years. Child sex and household head sex were measured as dichotomous variables, with a value of "1" if the child or the household head was male and "0" otherwise. The household size was measured as the number of individuals residing in each household. Location, which captures the household's geographical setting, was measured as a dichotomous variable, with a value of "1" for rural and "0" for urban households. Finally, children's school attendance was measured as a dichotomous variable, with a value of "1" if the child was currently attending school and "0" if not attending school.

This study uses a panel dataset to show the association between household shocks measured as parental death, climatic shock (floods or drought), and food price rise shocks. The Hausman specification test chooses the proper model between Random Effects (RE) and Fixed Effects (FE). In this study, we use the Hausman Specification Test; the results are presented in Table 1. The null hypothesis states that the preferred model has random effects due to its higher efficiency. The study adopted the random-effects approach when the p-value was more significant than 0.05.

Table 1: Hausman (1978) Specification Test Results.

Source: Own Computation, NPS (2014/15 &2019/20)

To estimate the effects of household shocks on the child-labour Equation, we use Equation (1):

Where:n_ijt represents the dependent variable, the child labour hours for individual in household j at time t The Shock_jt refers to the measure of household shocks. X represents a set of control variables, including individual and household characteristics θ is the household fixed effect, y_t is the survey wave fixed effect, and  is an error term. Household and child characteristics can influence the relationship between household shocks and child labour. Many of these factors are unobservable. Estimates that do not account for unobserved heterogeneity are biased. However, using panel data allows for controlling omitted variables even if these variables are unobserved [37]. This approach enables the generation of more reliable and accurate results on the effects of household shocks on child labour.

Summary Statistics

Table 2 presents the summary statistics of this study. A sample size of 588 individuals was used for the analysis, consisting of 47% male and 53% female children on average. The sample size was 588 individuals, averaging 47% male and 53% female. As shown in Table 2, the children worked an average of 28.55 hours per week. For children aged 12-14, working 28.55 hours per week was above the ILO's recommended limit for light work. An average age of around 50 years for household heads implies that many households are old, and about 80% of households are male-headed, suggesting that Tanzanian society is predominantly patrilineal. Married household heads account for approximately 74% of the sample population. Furthermore, the data indicate that approximately 70% of households reside in rural areas, with the remaining 30% living in urban areas. This implies that many children also live in rural areas where they work in child labour, including farming. On average, 88% of children aged 7-13 are enrolled in primary school. The average distance from home to the nearest school is approximately 21 minutes, indicating that children typically travel for approximately 21 minutes to reach school.

Table 2: Summary statistics.

Source: Authors' computations from NPS (2014/15 and 2019/20).

Effects of Household Shocks on Child Labour

Column [1] in Table 3 shows the estimations when shocks enter the regression equation without control variables, whereas Column [2] shows the estimations when shocks enter the regression equation with other control variables. From Table 3, climatic shocks such as floods or droughts significantly affect child labour hours. In Model [1], the coefficient is 0.869 (significant at the 5% level), suggesting that children work 0.869 more hours per week, on average, when a climatic shock occurs. In Model [2], the effect is slightly lower, with a coefficient of 0.621 (significant at the 10% level), but it still indicates that climatic shocks increase child labour hours. However, the effects of a food price rise and death shock are not statistically significant in either model, indicating that they do not have a discernible impact on child labour hours in this sample. Male children worked 0.722 more hours per week on average than female children (significant at the 5% level) compared to their female counterparts; for each additional year of age, children worked 0.440 more hours per week on average (significant at the 1% level). On average, children who attended school worked 0.971 fewer hours per week (significant at the 1% level). Children living in rural areas worked an average of 0.922 more hours per week than those living in urban areas (significant at the 5% level).

Individual Household Shocks on Child Labour

Table 4 presents the fixed-effects regression estimates of the effects of different household shocks on child labour when all shocks are included in one regression model. The analysis examines the individual correlation of each shock with the outcome variable of child labour. From Table 4, the coefficient for climatic shock is positive and significant at the 5% level in the model [1], indicating that climatic shock increases child labour hours by 0.611 on average. In contrast, the coefficient for the food price rise shock is positive and significant at the 1% level in the model [2]. This suggests that an increase in food prices leads to an average 0.308-hour increase in child labour hours. Exposure to death shocks in model [3] is negative but insignificant. Similarly, age increases with an increase in hours in all three models at the 1% or 5% levels, indicating that labour hours also increase as the child's age increases. Child school attendance decreased child labour hours in Models (1), (2), and (3). This suggests that children who attend school work fewer hours. These results suggest that children who attend school have fewer working hours. Further, the coefficient for rural location is positive and significant at the 1% level in models [2] and [3], indicating that children living in rural areas work more hours than those in urban areas. Column [1] of Table 4 results indicate that when all other factors remain constant and shocks enter the regression model individually, climatic shocks have a statistically significant positive effect on child labour at the 5 per cent significance level. This shows that climatic shocks are associated with significantly higher child labour hours. Specifically, when households experience climatic shocks, child labour increases by 0.61 hours per week. Furthermore, the results indicate that when a household experiences food price shocks (Column [2]), child labour hours increase by 0.30 per week. This implies that food price rise shocks have a statistically significant positive effect on child labour. When a household experiences food price shocks, child labour hours increase by 0.30 per week. The results in Column [3] do not show a statistically significant effect of death shocks on child labour hours, indicating that the death of a family member does not significantly impact child labour hours. It is important to note that other factors, such as child sex, age, school attendance, and location, remain significant in each regression model, suggesting that these factors consistently influence child labour hours across different household shocks.

Table 3: Effects of Household Shocks on child labour. (Fixed effects estimations).

Robust Standard errors in parentheses: *** p<0.01, ** p<0.05, *p<1.

Table 4: Household Shocks and Child labour: Fixed effects regression results (Individual shocks).

Robust Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1

prices, and climatic shocks, such as droughts and floods, can increase child labour. In these circumstances, children may be required to supplement family money frequently at the expense of their education and general welfare. The negative impacts of household shocks on child labour are particularly harmful to low-income households and those with less access to social safety nets. This study examines the impact of household shocks on child labour in Tanzania, specifically targeting children aged 7-13 enrolled in the Tanzanian education system. This study employed two waves of Tanzanian National Panel Survey data(2014/15 and 2019/20) to explore the link between household shocks measured by food price rise shocks, parental death shocks, and climatic shocks, including drought, floods, and child labour measured in hours. The results indicate that climatic shocks such as droughts and floods increase child labour hours. These findings align with those of Koohi-Kamali and Roy [34]. Mengstu [35] also observed a similar trend in Ethiopia, where child labour increased as drought conditions worsened. These findings suggest that households reliant on agricultural activities are vulnerable to income loss and decreased consumption levels owing to climatic shocks. Consequently, families may employ their children as coping mechanisms in response to drought, interpreting such events as signs that their consumption has fallen below subsistence levels [35]. Drought also affects child labour through the health and nutrition of affected children. Existing literature provides substantial evidence that drought leads to food insecurity, resulting in malnutrition among children. Consequently, these children may be forced to miss school and engage in work to support themselves and their families [36]. It is essential to recognize that many Tanzanian households rely on agriculture for their livelihood, an industry heavily influenced by climate fluctuations. As a result, parents whose income is dependent on agriculture might resort to involving their children in the labour market to compensate for lost income. A study conducted by Beegle [37] in Tanzania corroborates these findings, demonstrating that agricultural shocks, including rainfall deviations, cause an increase in child labour by one standard deviation (5.7 h). This increase ultimately results in a loss of approximately one year of education for affected children. The study further reveals that child labour hours increase as food prices increase. Food security may influence child labour in two ways. First, when food prices rise, households experience diminished purchasing power because of the inability of suppliers to rapidly augment food production [38]. As a result, food price shocks cause families to lose their actual income, reducing household purchasing power and causing food scarcity. In response, households may resort to involving their children in their work to supplement their family income, thereby enabling more efficient access to food. Death shocks had no significant effect on children's labour. In many societies, assistance from extended families is essential to overcome difficult times, including the shock of death. Family members frequently provide money or non-financial aid, such as childcare, which reduces the need for children to work [4]. Social protection systems, such as welfare or social insurance, can also help families deal with the financial strain caused by a death shock [39]. These initiatives offer financial assistance, allowing families to maintain their quality of life without turning to child labour.This study also investigates the link between controls and child labour. These findings indicate that child labour increases with age. There is evidence that children work longer hours as they age. A possible explanation for this trend is that older children are less likely to include their parents because they try to assert their independence. Therefore, young people may decide to work independently. On the other hand, the study discovered that child labour increases with a child's sex, suggesting that boys are more likely to work than their female counterparts. Gurmu and Tilahun [42] reported similar findings in Ethiopia. The ILO (2020) report in Tanzania also indicated that boys are more likely to engage in child labour than girls and various factors might cause boys to engage in child labour more often than girls. From a gender perspective and norms, boys are frequently expected to work and contribute to their family's income. In contrast, girls care for their homes and provide for others [42]. On the other hand, girls are frequently restricted to domestic work or care obligations that are not typically considered child labour. At the same time, males have more access to outside-the-home employment because of this dynamic [45]. Additionally, cultural beliefs and values may contribute to educational disparities because families frequently prioritize boys' education over girls', leaving daughters at home to care for while their brothers’ work to support the family [44]. Furthermore, child school attendance was strongly negatively correlated with child labour hours. This indicates that as children attend school, their involvement in child labour tends to diminish. Education giving children alternative alternatives and a road to a better future may be one explanation for this correlation, which would reduce the amount of child labour they engage in [15]. According to earlier research, children from wealthier families and those with household heads with formal education are more likely to attend school. They are less likely to engage in child labour than children from low-income families and families without formal education [15]. This is explained by the fact that families with stable finances can better invest in their children's education and depend less on child labour to make ends meet. Moreover, household heads with higher levels of education are more likely to recognize the long-term advantages of education and place a higher value on their children's education than on the short-term financial gains from child labour. Previous studies have found that children from wealthier homes and those whose heads have formal education are more likely to attend school. These children are also less likely to engage in child labour than those from families with no formal education or low-income households [15]. Finally, this study shows that child labour affects children residing in rural areas more than their urban counterparts. In the Tanzanian context, child labour may persist in rural areas regardless of a family's wealth. It may be challenging to gauge a family's financial situation in rural areas because revenue sources are frequently sporadic or dependent on the season. Additionally, social services and educational opportunities may be scarce in rural areas, which can make child labour more pervasive. Owing to cultural conventions, a lack of educational possibilities, or the necessity of helping with household survival activities, even children from reasonably wealthy households may be forced to work in these situations. Devi and Roy [47] also supported this finding.

This study on child labour in Tanzania concluded that household shocks, such as climatic shocks and increased food prices, significantly increased the number of hours children worked. The results highlight the necessity of implementing shock-mitigation measures to safeguard vulnerable households, such as creating drought-resistant crop types and small-scale irrigation systems. School attendance for children should be implemented to reduce their impact on child labour. In addition, policies should consider the improvement of school attendance for children, consider various home socioeconomic circumstances, and handle unexpected increases in food prices. Further, policies should create specific initiatives to enhance the environment for self-employed households and plan suitable responses to lessen dependency on child labour during shocks. The government and stakeholders may endeavor to reduce the effect of household shocks on child labour and promote a more secure future for Tanzanian children. Further, although not mentioned in the analysis, the study also recommends that households adopt shock-mitigating approaches, such as enhancing irrigation infrastructure and developing drought-resistant crop varieties, which will help them cope with climatic shocks rather than withdraw their children to school and engage them in child labour activities to compensate for income loss during household shocks. Lastly, potential biases in the data, a lack of generalizability, and failure to account for certain confounding factors are just a few of the limitations of the paper. Future studies could evaluate the efficacy of various policy interventions, investigate the role of cultural norms and gender dynamics in influencing child labour practices, and examine the long-term effects of household shocks on children's well-being.

1. Hyder A, Behrman JR and Kohler HP. Negative economic shocks and child schooling: Evidence from rural Malawi. Development Southern Africa. 2015; 32: 458-476.
2. Onisanwa ID and Olaniyan O. Health shocks and consumption smoothing among rural households in Nigeria. Journal of Economics & Management. 2019.
3. Skoufias E and Vinha K. Climate variability and child height in rural Mexico. Economics & Human Biology. 2012; 10: 54-73.
4. Bhalotra S and Heady C. Child farm labour: The wealth paradox. The World Bank Economic Review. 2003; 17: 197-227.
5. Guarcello L, Kovrova I and Rosati FC. Child labour as a response to shocks: evidence from Cambodian villages. Understanding Children's Work Project Working Paper Series. 2008.
6. Baez JE and Santos IV. Children's vulnerability to weather shocks: A natural disaster as a natural experiment. Social Science Research Network, New York. 2007.
7. Smith MM, Saklofske DH, Keefer KV and Tremblay PF. Coping strategies and psychological outcomes: The moderating effects of personal resiliency. The journal of psychology. 2016; 50: 318-332.
8. Beegle K, Dehejia RH and Gatti R. Child labour and agricultural shocks. Journal of Development Economics. 2006; 81: 80-96.
9. Jensen R. Agricultural volatility and investments in children. American Economic Review. 2000; 90: 399-404.
10. Skoufias E. Economic crises and natural disasters: Coping strategies and policy implications. World development. 2003; 31: 1087-1102.
11. Maccini S and Yang D. Under the weather: Health, schooling, and economic consequences of early-life rainfall. American Economic Review. 2009; 99: 1006-1026.
12. Huang WC and Acheampong K. Household Shocks and Child Labour: Evidence from a Panel Survey of Nigerian Households. 2018.
13. Sharma H and Gibson JK. Effects of International Migration on Child Schooling and Child Labour: Evidence from Nepal. University of Waikato.2020.
14. Khanam R. impact of child labour on school attendance and school attainment: Evidence from Bangladesh. In Population Association of America Annual Meeting (PAA 2005). University of Southern Queensland. 2006.
15. Huebler F. Child labour and school attendance: Evidence from MICS and DHS surveys. In Seminar on child labour, education and youth employment Understanding Children's Work Project Universidad Carlos III de Madrid Madrid. 2008; 11-12.
16. David SAHN, Glick P and Walker TF. Household shocks and education investment in Madagascar. 2012; 201227.
17. Duryea S, Lam D and Levison D. Effects of economic shocks on children's employment and schooling in Brazil. Journal of Development Economics. 2007; 84: 188-214.
18. Edmonds EV. Child labour. Handbook of development economics. 2007; 4: 3607-3709.
19. Bandara A, Dehejia R and Lavie-Rouse S. The impact of income and non-income shocks on child labour: Evidence from a panel survey of Tanzania. World development. 2015; 67: 218-237.
20. Delap E. Economic and cultural forces in the child labour debate: Evidence from urban Bangladesh. Journal of Development Studies. 2001; 37: 1-22.
21. Woldehanna T, Jones N and Tefera B. The invisibility of children's paid and unpaid work: implications for Ethiopia's national poverty reduction policy. Childhood. 2008; 15: 177-201.
22. D'Souza A and Tandon S. Intrahousehold nutritional inequities in rural Bangladesh. Economic Development and Cultural Change. 2019; 67: 625-657.
23. Dumas C. Why do parents make their children work? A test of the poverty hypothesis in rural areas of Burkina Faso. Oxford economic papers. 2007; 59: 301-329.
24. Webbink E, Smits J and de Jong E. Household and context determinants of child labour in 221 districts of 18 developing countries. Social indicators research.110; 819-836.what can we do? The World Bank Economic Review. 2013; 17: 147-173.
25. Skoufias E. Economic crises and natural disasters: Coping strategies and policy implications. World development. 2003; 31: 1087-1102.
26. Edmonds E. Does Child Labor Decline with Improving Economic Status? Journal of Human Resources. 2005; 40: 77-99.
27. Basu K. Child Labor: Cause Consequence and Cure with Remarks on International Labor Standards. Journal of Economic Literature. 1999; 37: 1083-1119.
28. Basu K. Child labour: cause consequence and cure with remarks on international labour standards. Journal of Economic Literature. 1999; 37: 1083-1119.
29. Ennew J and Swart-Kruger J. Researching Children’s Experiences: Methods and Methodological Issues. UNICEF. 2003.
30. Bhagwati J. In Defense of Globalization. Oxford University Press. 2004.
31. Becker GS. Human Capital: A Theoretical and Empirical Analysis, with Special Reference to Education. University of Chicago Press. 1964.
32. Bandura A. Social Learning Theory. Prentice-Hall.1977.
33. Edmonds EV. Child labour. Handbook of development economics. 2007; 4: 3607-3709.
34. Koohi-Kamali F and Roy A. Environmental Shocks and Child Labour: A Panel Data Evidence from Ethiopia & India. 2021.
35. Mengstu KA. The effect of drought on child labour in Ethiopia(Master's thesis). 2017.
36. Frempong RB and Stadelmann D. The effect of food price changes on child labour: Evidence from Uganda. The Journal of Development Studies. 2019; 55: 1492-1507.
37. Beegle K, Dehejia RH and Gatti R. Child labour and agricultural shocks. Journal of Development Economics. 2006; 81: 80-96.
38. Yousif IEAK and Al-Kahtani SH. Effects of high food prices on consumption pattern of Saudi consumers: A case study of Al Riyadh city. Journal of the Saudi Society of Agricultural Sciences. 2014; 13: 169-173.
39. Skoufias E. Economic crises and natural disasters: Coping strategies and policy implications. World development. 2003; 31: 1087-1102.
40. Skoufias E, Parker SW. Labour market shocks and their impacts on work and schooling: evidence from urban Mexico. 2002; 583: 2016-39587.
41. Skoufias E and Vinha K. Climate variability and child height in rural Mexico. Economics & Human Biology. 2012; 10: 54-73.
42. Kambhampati US and Rajan R. The 'nowhere children: Patriarchy and the role of girls in India's rural economy. The Journal of Development Studies. 2008; 44: 1309-1341.
43. Kondylis F and Manacorda M. School proximity and child labour evidence from rural Tanzania. Journal of Human Resources. 2012; 47: 32-63.
44. Edmonds EV. Child labour. Handbook of development economics. 2007; 4: 3607-3709.
45. Edmonds E. Does Child Labor Decline with Improving Economic Status? Journal of Human Resources. 2005; 40: 77-99.
46. Edmonds EV. Economic Growth and Child Labour in Low-Income Economies. A Systhesis Paper Prepared for IZA/DFID. Bonn: Institute for the Paper of Labour. 2015.
47. Devi K and Roy G. Study of child labour among school children in urban and rural areas of Pondicherry. Indian Journal of Community Medicine: Official Publication of Indian Association of Preventive & Social Medicine. 2008; 33: 116.
48. Baland JM and Robinson JA. Is Child Labour Inefficient? Journal of Political Economy. 2000; 108: 663-679.
49. Ennew J and Swart-Kruger J. Researching Children’s Experiences: Methods and Methodological Issues. UNICEF. 2003.
50. Guarcello L, Kovrova I and Rosati FC. Child labour as a response to shocks: evidence from Cambodian villages. Understanding Children's Work Project Working Paper Series. 2008.
51. Khanam R. impact of child labour on school attendance and school attainment: Evidence from Bangladesh. In Population Association of America Annual Meeting (PAA 2005). University of Southern Queensland.2006.
52. Sen A. Gender and cooperative conflicts. In I. Tinker (Ed.) Persistent inequalities: Women and world development Oxford University Press. 1990; 123-149.
53. United Republic of Tanzania. National Population and Housing Census. Dar es Salaam: National Bureau of Statistics Ministry of Finance.2022.
54. Yousif IEAK and Al-Kahtani SH. Effects of high food prices on consumption pattern of Saudi consumers: A case study of Al Riyadh city. Journal of the Saudi Society of Agricultural Sciences. 2014; 13:169-173.
55. Woldehanna T, Jones N and Tefera B. The invisibility of children's paid and unpaid work: implications for Ethiopia's national poverty reduction policy. Childhood. 2008; 15: 177-201.
56. Jensen R. Agricultural volatility and investments in children. American Economic Review. 2000; 90: 399-404.