Global literature on child marriage is heavily skewed towards social, demographic and health implications. In particular, most of these studies have reported differences in health, education and employment outcomes for girls married before and after turning 18 years old. Few studies have documented the economic cost of child marriage, particularly the macroeconomic costs (e.g. percentage of GDP loss due to child marriage). The literature constraint is more significant for the Arab region than other regions of the world; thus, we have utilized both global and Arab region studies to build conceptual and analytical frameworks for the current study. We have organized the literature review into three sections: (1) drivers of child marriage; (2) economic costs; and (3) child marriage in the context of the Arab region.

Drivers: The documentation on child marriage consistently highlights several structural factors that generate and intensify child marriage, ranging from economic factors such as poverty and limited work opportunities; sociocultural factors such as education, social practices, religious beliefs, ethnicity, class and gender norms; and political factors such as instability, including conflict, displacement and natural disasters. All these factors jeopardize a girl’s voice and autonomy, putting her in the loop of early marriage and its eventual consequences. The poorest countries, regions and households often confront the greatest prevalence of child marriage, with poor girls residing in rural regions as the most vulnerable. Moreover, the lack of work opportunities for girls due to social norms and practices may cause parents to consider it unnecessary to invest in their schooling.

More often than not, as the growing evidence shows, the level of education tends to determine a girl’s age at marriage; as such, lower education attainment is associated with a lower age at marriage. Furthermore, child marriage has been demonstrated to be deeply entrenched in social practices and traditions. Diverse settings serve as drivers for child marriage that are prevalent in respective set-ups, notably the practice of dowry or bride wealth, which can generate instant economic gains for a family; community pressure to conform to societal norms; using the marriage of girls to settle family disputes; the fear of sexual harassment or sexual violence; the desire to control girls’ sexuality to avoid unwanted pregnancies or jeopardizing the family’s honour; and internalized social norms whereby girls themselves desire to marry early due to perceived vulnerabilities and a lack of alternatives.

Conflict and war impact women and girls in uniquely gendered ways, albeit varying based on the location, earnings, social set-up and cultural setting. Conflict-induced instability generates fear of injury and death, escalates incidents of sexual violence, triggers food insecurity and deepens gender stereotypes. Such instability leads to a low mean age for females at the time of marriage, higher child marriage rates and low female literacy rates. For families in conflict zones, child marriage becomes a negative coping mechanism to “save” the girl from perceived exploitation while conserving limited resources by passing responsibility for her to another household. Child marriage, therefore, can be interpreted as a social exchange of girls by the family to maximize their resources and safety nets.

The economic costs of child marriage: Age at marriage is a significant determinant of population dynamics, given that it sets the foundation for forthcoming factors in deciding a girl’s quality of life. While child marriage is widely addressed as a human and women’s rights issue, lately, studies have highlighted and quantified the economic costs of child marriage (table 1).

The economic cost of violence against women and girls (child marriage, in this case), defined in the UN-Women report “The costs of violence”, is the direct and indirect tangible cost with a monetary value. These could be the private costs endured by young girls and their next generation, or public costs such as an increased burden on the government’s health-care and education systems. The total costs tend to have a multiplier effect on GDP and economic development, triggering the vicious cycle of inter-generational poverty and inequality.

Child marriage directly lowers women’s work prospects and financial returns due to low educational levels. At the same time, it indirectly increases the proportion of their unpaid household work resulting from higher lifetime fertility. To corroborate, Savadogo and Wodon (2017a) found that child marriage reduces earnings in adulthood for women marrying early by 9 per cent through its impact on education. In some countries, it has been noted to affect decision-making and bargaining power.

Therefore, economic costs are the closest channels affecting the GDP of a State through the low employment rate of women married as children, the low wage rate of their unskilled jobs, and lower earnings, savings and hence a lower tax generation for the State.

Child marriage in Arab countries: Like many other regions in the world, the Arab region has patriarchal norms whereby women are expected to prioritize their family before their own rights as individuals.

This includes the institutionalization of policies that work to preserve the patriarchal status quo that ultimately governs economic, political and social decision-making.

In the past decade, conflict has tended to drive the many instances of child marriage in the region, resulting in refugees and displaced families resorting to the practice to protect girls from sexual violence and thus safeguard the family’s honour.

In some cases, child marriage is linked to kidnapping and trafficking by armed groups and militias; the Syrian Arab Republic and Iraq have recorded such instances of the abduction of girls.

In Jordan, child marriage incidents have increased since the onset of the conflict in the Syrian Arab Republic in 2011, particularly among Syrian refugees. For example, the rate of child marriage among Syrian girls in Jordan increased from 33.1 per cent in 2010 to 43.8 per cent in 2015, impacting their sexual and reproductive health in terms of untimely pregnancies, domestic violence, social alienation, mental health issues and a loss of work opportunities.

There is strong evidence of the relationship between child marriage and deep-rooted cultural beliefs and discriminatory gender norms in the Arab region. For instance, in Egypt, child marriage is linked with community notions related to female genital mutilation (FGM), while in the Syrian Arab Republic, young girls are persuaded to marry at a young age due to their sexual inexperience.

Poverty is another factor that engenders child marriage in countries such as Egypt, Libya, Somalia, the Sudan, the Syrian Arab Republic and Yemen.

The trend is also noted in Iraq, Jordan and Morocco, where girls in low-income families, viewed as a financial burden, were twice as likely to marry young than in wealthier households in 2006–2011.

A study on Syrian refugees in Egypt highlights that underperforming girls in school were better off married, and girls interested in education were retained in school to continue their studies and had their marriages postponed.

Among other factors are a worsening economy and growing inflation rates that negatively impact the survival of low-income families, particularly with young women, in these regions.

Phase I of this exercise explored the costs of child marriage on women and girls in four Arab countries and is the only comprehensive study that has highlighted the social and health costs of child marriage at different stages of women’s lives. Women and girls who are married young in Egypt, Iraq, Jordan and Tunisia experienced serious ramifications at each stage of life in terms of fertility, decision-making, education, autonomy, labour force participation and mortality rates of their children. Hence, Phase II builds on these findings to focus on the loss of income (or GDP) for the Arab region.

In the Arab region, research has mainly focused on examining the prevalence of child marriage, identifying the social, economic and political factors that drive this practice, and investigating its harmful effects on girls and women. While the Phase I report established a link between child marriage and various outcomes related to demographics, health, education and the labour market, it did not quantify the economic losses that result from this practice for the State. To gain a better understanding of the actual short- and long-term economic costs that women, their families and the Government will bear, it is crucial to evaluate the impact of child marriage on the GDP of all Arab countries.

Therefore, the goal of this study is to conduct a cost analysis that measures the impact of child marriage on the GDP of countries, with the aim of strengthening the case for eradicating this practice in the region. The study advances recommendations to reduce the economic burden of child marriage by taking proactive steps to address the intermediate channels that contribute to economic implications.

The present study employs the spectrumbased simulation model and incorporates a wide range of input indicators to provide a detailed analysis of the demographic, social and macroeconomic costs associated with child marriage. Unlike Mitra and others (2020), the present study does not rely solely on a macroregression model approach. Instead, its hierarchical and component simulation approach enables the inclusion of more input parameters and the estimation of a larger number of outcome indicators. Moreover, the model used in the study not only projects the cost of inaction on child marriage as a percentage of GDP loss but also includes other key demographic and health indicators up to 2050, covering the entire productive lifespan of a girl who was married at age 15 around the year 2000.

Comparatively, the Arab region lacks the availability of reliable and consistent data to measure and monitor child marriage and its causes and consequences. Recurrent conflicts and geopolitical issues in the region worsen the global periodic surveys critical for cross-country comparisons. Of the 22 countries that constitute the Arab region, only 13 countries had relevant data for the time period under consideration (2001–2020). For the exercise of costing child marriage in this region, the data for the input indicators was collected and compiled from multiple sources for the period 2001–2020. The data sources used for analysis include: (i) demographic health surveys (DHS); (ii) multiple indicator cluster surveys (MICS); (iii) labour force surveys (LFS); (iv) United Nations World Population Prospects; (v) world development indicators (WDI); (vi) United Nations Model Life Table (West Asia Model); (vii) countryspecific censuses from Arab countries; and (viii) official statistics of the respective Arab countries. The present study adopted interpolation and extrapolation methods to fill the data gaps between any years.

Globally, the DHS is the largest data source for population, health, human immunodeficiency virus (HIV) and nutrition and is internationally comparable and surveyed in about 90 countries. Some of the Arab countries have been covered under the DHS programme, such as Egypt (2014, special DHS in 2015), Jordan (2017/2018), Mauritania (2019–2021), Morocco (2003/2004), the Sudan (1989/1990), Tunisia (1988) and Yemen (2013). The MICS is the largest source of reliable internationally comparable data on women and children globally and has been carried out once or more in 118 countries. The latest MICS within the Arab region are: Algeria (2018/2019), Egypt (2013/2014), Iraq (2018), Mauritania (2015), Oman (2014 restricted), the State of Palestine (2018/2019), Qatar (2012), Somalia (2011), the Sudan (2014), the Syrian Arab Republic (2006), Tunisia (2018) and Yemen (2006). Although the Pan Arab Project for Family Health (PAPFAM) is available in Libya, the study excluded Libya, due to a lack of information comparability in PAPFAM with the DHS and MICS, and Oman, due to a lack of access to the microdata. Overall, the study covers 13 Arab countries that have microdata from the latest available DHS or MICS.

In addition, we collected information on employment and unemployment from the WDI, LFS and official statistics of the respective countries. The total and projected population data were collected from the countries’ censuses and United Nations World Population Prospects, respectively. The study used the most suitable model life table from two sets of standard model life table families to derive a variety of mortality indicators and underlying mortality patterns for the estimation and projection of the population for each country. The data on the base year GDP in United States dollars ($), annual GDP growth rate (%), and urbanization (%) were collected from the WDI. In addition, education and health-related indicators were compiled from multiple data sources including DHS, MICS, censuses, and official statistics of the countries (see the details in annex table 1).

The linkage between child marriage and economic growth is not straightforward because it directly correlates with some “conventional” economic growth determinants such as fertility, education, health and employment, among others. Wodon and others (2017) outlined five main channels – health, education, fertility, labour force participation and decisionmaking – through which child marriage impacts economic growth. In their analytical model, Wodon and colleagues further consolidated these channels into human capital (i.e. education and health), as there is a significant overlap across these channels. Later, the study by Mitra and others (2020) also recognized the interdependence between health, education, economic growth and other intermediates operating under the costing exercise of child marriage.

Following Wodon and others (2017) and Mitra and others (2020), this study used four sets of parameters (i.e. demographic, health, education and economic) in the costing exercise of child marriage in a life course perspective. Child marriage is like a silhouette on lifetime outcomes, as it affects skill formation, health and economic consequences at all stages of life. In this study, by saying “life course”, we mean the estimates are cumulative economic costs of child marriage associated with education, health and labour market losses, which operate at different stages of an individual life (box below). Moreover, several other intermediates – such as women’s decision-making ability and gender-based violence against women – that influence economic outcomes were not included as separate variables because they are highly collinear with demographic, health, education and employment parameters. Nevertheless, within four broad sets of parameters (demographic, health, education and economic), the model used in this study includes other contingent factors such as age structure of the population and its drivers, economic status and its predictors, and the urbanization level of women’s country of residence. Figure 2 explains the complete operational (or analytical) framework of the simulation model and the parameters used to estimate the economic costs of child marriage in the Arab region.

While the present study follows the simulation approach used by Wodon and others (2017), its analytical framework for estimating the economic costs of child marriage is slightly different and adheres to a more comprehensive procedure. We have extended the projection of the economic costs of child marriage for the Arab region up to 2050 with the base year of 2001. The period 2001 to 2050 is selected considering the working lifespan of around 50 years for a female married at the age of 15. For instance, in this model, a female who married in the year 2001 at age 15 (base year for this study) is expected to live up to 65 years (up to 2050, a goalpost year for this study) as a worker or a non-worker. However, the input indicators are not available for all countries from 2001. In such cases, the base year has been chosen as per the availability of the data. The life-cycle approach.

The FamPlan module stands for the projection of family planning parameters. Family planning inputs are needed to reach national goals for addressing unmet needs or achieving desired fertility. For this study, the family planning module provides necessary parameters that can predict probable differences in family planning indicators and their consequences for the fertility of child-married women and non-child married women.

Resources for the Awareness of Population Impacts on Development (RAPID) projects the social and economic consequences of high fertility and rapid population growth for such sectors as labour, education, health, urbanization and agriculture. For the simulation model, RAPID provides the differential probability of socioeconomic achievements of child-married and nonchild married input and outcome indicators. A detailed explanation of these modules is presented in annex table 2.

The study also differs in terms of its outcome measures. We have provided macro-level demographic and health costs alongside the economic costs (GDP loss) for three different scenarios. These three scenarios include: (i) child marriage scenario – a hypothetical case where we assume all women across the Arab region are married below the age of 18; (ii) non-child marriage scenario – the best hypothetical case where we assume that all women across the Arab region are married at 18 years of age or above; and (iii) overall scenario (as usual scenario) – a case where the status quo continues, that is, child marriage continues to prevail at the current level in the Arab region.

As the most populous country in the Arab region, Egypt contributes around 24 per cent of the population. At the country level, child marriage prevalence stands at 17.4 per cent as at 2014. However, there are considerable variations among the provinces, with Fayoum, Beni Suef and Giza showing as high as 27 per cent, 26 per cent and 25 per cent, respectively, and Suez showing just 4.4 per cent of females married before the age of 18.

Iraq constitutes 9 per cent of the total population of the Arab region. In terms of the prevalence of child marriage, with 28 per cent of women married before 18 years of age, Iraq stands among the top five countries regionally. Moreover, the sub-national pattern indicates a significant variation across the 18 provinces. The child marriage rate is lowest in Duhok with 8.14 per cent and highest in Misan with 44 per cent. Najaf (37.2 per cent), Karbala (37 per cent), Thi-Qar (35 per cent), Basrah (33.5 per cent), Diala (32 per cent) and Nineveh (31 per cent) exhibit more than 30 per cent of women marrying before the age of 18, as at 2018.

With 9.7 per cent of females married below 18 years of age at the national level in 2017/2018, Jordan shows a moderate prevalence of child marriage in the region. Jordan also shows a considerable variation in the sub-national pattern of child marriage prevalence. Across the 12 governorates in the country, the child marriage rate ranges from as low as 3.27 per cent in Tafilah to 15.4 per cent in Mafraq.

In terms of the prevalence of child marriage, with 36.6 per cent of females married before the age of 18, Mauritania stands in second place among all the Arab countries considered for the study. Besides, the subnational pattern also indicates a significant variation across the 14 regions. The child marriage rate is lowest in Nouakchott-Ouest at 16.5 per cent and highest in Guidimaka at 57.3 per cent. Gorgol (50.3 per cent), Hodh Ech Chargui (49.7 per cent), Assaba (46.2 per cent) and Hodh El Gharbi (43.7 per cent) depict more than 40 per cent of females marrying before age 18 as of 2019–2021.

The average child marriage rate in Morocco is 16 per cent as at 2003/2004. Within the country, the percentage of females married before the age of 18 varies from 7.7 per cent in Grand-Casablanca to 29.4 per cent in Laâyoune-Boujdou-Sakia Al Hamra.

With 13.4 per cent of females married below 18 years of age at the national level in 2019/2020, the State of Palestine shows a moderate prevalence of child marriage in the region. The State of Palestine also indicates a substantial difference in the sub-national pattern of child marriage prevalence. Across the 16 governorates, the child marriage rate ranges from as low as 5.3 per cent in Tulkarem to 22.8 per cent in North Gaza.

With an average of 4.2 per cent of females marrying before they are 18 years old in 2012, Qatar shows one of the lowest levels of child marriage prevalence in the Arab region. However, within the country, the percentage of females married before age 18 varies from 0 per cent in Al-Shamal and Al-Wakra to 13.36 per cent in Al-Daayen.

The average prevalence of child marriage in Somalia is 45.3 per cent, which is the highest among all the Arab countries included in this study. Across the regions, the prevalence of child marriage varies considerably. It ranges from 12 per cent in Awdal to 90 per cent in Middle Juba. Ten out of 18 regions in the country have more than the national average (i.e. 45 per cent and above).

The Sudan is the second largest country, contributing around 10 per cent of the population of the Arab region. At the country level, the child marriage prevalence rate in the Sudan stands at 34.2 per cent as at 2010. However, there are considerable variations in child marriage prevalence across 18 states, ranging from 56 per cent in Central Darfur to 17.6 per cent in River Nile. South Darfur (52.3 per cent), Blue Nile (49.6 per cent), El Gadarif (47.5 per cent), East Darfur (46.3 per cent), West Darfur (45 per cent), South Kordofan (43.7 per cent) and Kassala (40.6 per cent) have over 40 per cent of females marrying before they are 18 years old.

With an average of 13.3 per cent of females marrying before 18 years as at 2012, the Syrian Arab Republic shows a moderate child marriage prevalence in the Arab region. However, within the country, the percentage of females married before the age of 18 varies from 4.7 per cent in Tartus to 26 per cent in Quneitra.

At the country level, with an average child marriage rate of 1.5 per cent as at 2018, Tunisia shows the lowest prevalence among the 13 Arab countries included in this study. Even within the country, the percentage of females married before the age of 18 does not vary significantly.

With an average of 32 per cent of females marrying before 18 years as at 2013, Yemen stands in the top five countries in terms of child marriage prevalence in the Arab region. Also, the country shows huge sub-national variation across its 21 governorates. The percentage of females married before age 18 varies from 50.5 per cent in Dhamar to 10 per cent in Aden. Besides Dhamar, three other governorates (Al-Jawf, Al-Mhweit and Raymah) show a child marriage prevalence rate of more than 40 per cent.

Education is a key human capital measure that predicts economic growth.45 Disruptions in the educational attainment of childmarried women are widely recognized in the literature. Females married below 18 years of age are less likely to enter higher education compared to women married at higher ages.46 Lower educational levels for women bring poor social and economic outcomes not only to the individual but to the respective households and the State as well.47 Thus, education is a mechanism through which child marriage may affect the State economy. Table 4 presents educational differences by age at first marriage. Findings suggest considerable but varying differences in higher education across women married below 18 years of age and those above 18 years of age. For instance, in countries like Algeria and Morocco, the difference in higher education among females married below 18 years of age and those who married at the age of 18 and above is 20 and 67 times, respectively, while the differences are the least in Qatar (2 times) and Iraq (5 times). Other countries, such as Egypt, Somalia, the Sudan, the Syrian Arab Republic and Yemen, show higher educational attainment ranging among women married at and above 18 years of age compared to females married below 18 years of age. Differences in the education levels for Jordan, Mauritania, the State of Palestine and Tunisia are around seven times on average, with women married at the age of 18 and above faring better in all these countries.

The workforce participation rate of a population has a direct bearing on a household and country’s economic prospects, while child marriage has an effect on the individual’s participation in the labour market.48 Even so, a male’s labour force participation increases after marriage while a female’s declines. The ILOSTAT, using data from 107 countries, has revealed that men have a higher level of participation in the labour force than females, while this gender gap is worsened for married men and women.49 Furthermore, childbearing also decreases the participation among females.50 Table 5 presents the differences in participation in the labour market by married females below 18 years of age and women married at 18 years and above for those in the age group of 15-49 years for five Arab countries. The findings suggest considerably low but varying differences in the workforce participation rate for females married below 18 years of age compared to those married at 18 years and above. For instance, the workforce participation rate for females married below 18 years of age in Algeria is nearly three times less with reference to those married at 18 years and above. Even in Morocco and Egypt, the differences in women’s workforce participation rate are considerably low for females married below 18 years of age compared to those married at 18 years and above. Furthermore, Mauritania and Yemen also exhibit a lower workforce participation rate for females married below 18 years of age compared to those married at 18 years and above.

Table 6 shows the percentage of GDP lost due to child marriage across the 13 Arab countries included in this study. In 2021, Algeria lost the highest percentage of GDP (5.1 per cent), while Qatar lost the lowest percentage of GDP (0.1 per cent). Ten countries lost over 2 per cent of their GDP, with the State of Palestine, the Sudan and Tunisia losing over 4 per cent (4.3 per cent, 4.9 per cent and 4.3 per cent, respectively). The remaining three countries had lower than 2 per cent of their GDP lost due to child marriage, with only Qatar and the Syrian Arab Republic below 0.5 per cent (0.1 per cent and 0.3 per cent, respectively).

It is important to note that the economic cost of child marriage not only depends on the rates of child marriage but also on the differences in demographic and socioeconomic outcomes between females married below 18 years of age and those married at 18 years and above. Therefore, Algeria, Jordan and Tunisia will incur greater economic costs attributable to child marriage because they have greater fertility and educational differences across females married below 18 years of age than those married at 18 years and above. While, for countries like Iraq and Mauritania, despite having higher child marriage prevalence rates, the relative differences in fertility rates and education levels between those married below 18 years of age and those married at 18 years and above are not as high as those in Algeria, Jordan and Tunisia, thus incurring lesser economic costs that are attributable to child marriage.

Given the model used in this costing exercise, the trends remain more or less the same in the projected years. In 2050, the Sudan is projected to lose the highest percentage of its GDP (5.1 per cent), while Qatar is projected to lose the lowest at 0.1 per cent. Eight countries are projected to lose over 2 per cent of their GDP due to child marriage, with the average being 4.06 per cent. The remaining five countries – Iraq, Mauritania, Qatar, Somalia and the Syrian Arab Republic – are forecasted to lose less than 2 per cent of their respective GDP, implying that good socioeconomic and health resources could help avoid the high economic costs of child marriage.

The estimated economic cost of child marriage for the Arab region stood at 3.2 per cent in 2021 and will be 3 per cent in 2050.

As explained through the conceptual framework of this study, a part of the economic cost of child marriage for countries influences the household as well. Households with a woman married as a child experience greater income loss both through wages as well as excess health-care expenditures. Earlier, Wodon and others (2017), as well as Wodon and Yedan (2017b), demonstrated that the economic participation and wage earnings for females married at early ages are significantly less compared to their counterparts married at higher ages. Wages have a significant bearing on household earnings.

This section provides an analysis of the average annual economic costs borne by households and the private health-care costs associated with child marriage. The findings presented in figure 5 indicate that households in most Arab countries face significant average annual economic costs. For example, seven Arab countries (Algeria, Egypt, Iraq, Jordan, Morocco, the State of Palestine and Tunisia) show average economic costs exceeding $600 per household, with Algeria having the highest cost at $1,173 and the Syrian Arab Republic having the lowest at $37.

Child marriage has been shown to have lifetime consequences for girls regarding poor educational, health and economic outcomes, depriving them of fundamental rights and leaving the next generation at a disadvantage. The issue has been aggravated by the COVID-19 pandemic and demands effective and conscious intervention, especially in the poorest countries, which exhibit the highest rates of child marriage. A growing volume of studies are increasingly demonstrating the harmful effects of child marriage on varied aspects of development outcomes. However, concerted efforts and resources to neutralize the practice remain inadequate across the Arab region. To stimulate efforts towards ending child marriage, the present study presents the economic costs of child marriage and its key mechanisms.

The respective economic costs of child marriage for 13 Arab countries have been estimated from 2001 to 2050. However, the timeline varies for a few countries due to the non-availability of data. The estimated economic cost of child marriage in terms of total GDP and the percentage of GDP lost provide compelling evidence that child marriage induces enormous and exponential economic costs for the Arab region. The GDP lost due to child marriage across 13 Arab countries is estimated at 3.2 per cent for 2021 and projected to be 3 per cent in 2050, with the cumulative GDP loss around $3 trillion during the forecasted period. Country-wise, Algeria and the Sudan were estimated to lose the highest percentage of GDP due to child marriage in 2021, while Qatar lost the lowest rate of GDP.

On the other hand, in 2050, the Sudan (5.1 per cent), Algeria (4.8 per cent) and Tunisia (4.6 per cent) are projected to lose the highest GDP due to child marriage if the current rate persists. The country-level differentials in the economic costs of child marriage in terms of GDP loss are both due to the level of child marriage and endowment factors such as the quality of health care and the socioeconomic system. It is possible that, despite similar child marriage rates, some countries have managed to control the damage caused by child marriage through better health-care and socioeconomic systems, thus reflected in the lower percentages of GDP loss. And some countries, such as the Sudan, have not experienced a significant difference in demographic and socioeconomic outcomes between females married below 18 years of age and those married at 18 years and above. Thus, economic costs solely attributable to child marriage are relatively less despite having a higher prevalence of child marriage and inferior demographic and socioeconomic outcomes in the country.

At the outset, the study finds that the economic cost of child marriage is substantial across the Arab region. Our estimate (3.1 per cent in 2021 for the Arab region) is slightly on the higher side compared to the 1.05 per cent reported by Mitra and others (2020) in the case of EMDCs and the 1.44 per cent noted by Wodon and others (2017) in the case of South Asian, Middle Eastern and African countries. The higher side estimate from the current study can be attributed to a greater number of components (i.e. direct and indirect costs) considered for the estimation. The estimated economic cost of child marriage in this study accounts for several direct and indirect costs, as shown in the analytical framework (figure 2). The economic cost due to child marriage not only depends on the level of its prevalence but also on the economies of scale and the country’s socioeconomic, demographic and health policies.

From a policy perspective, the study suggests that Arab countries can increase their GDP around 3 per cent by eliminating child marriage. It is important to note that the current study does not engage in detailed empirical analyses of all the pathways through which child marriage impacts the economy of a State and intervention strategies to eliminate child marriage because they are widely documented in the existing literature. Along with its conceptual framework, a synthesis of the empirical evidence found in this study in the context of previous literature provides insights into the mechanisms through which child marriage induces economic costs for a State and also provides possible intervention strategies to overcome this consequence. The critical mechanisms identified are demographic, social and health implications. Demographic implications comprise unwanted pregnancies and unsafe abortions that alter future growth, survival and/or reproduction. Social implications include the loss of educational attainment by girls who are married as children, which eventually harms the exercise of their basic rights, agency, decision-making ability, earning prospects, community support and empowerment in general. Health implications include the high fertility rates of females marrying early and higher maternal morbidity and mortality rates. These implications, endured by girls who marry early, might be direct or indirect, as well as monetary or non-monetary, for individuals and households and are cumulated at the State level.

In conclusion, the present study advances the suggestions put forward by Asha George and others (2020) that eliminating child marriage also requires addressing the structural determinants of gender inequality. The sooner a Government acts to eradicate child marriage, the greater the economic savings. Although the financial costs should not be the only reason for investing in the end of this practice, it certainly is a paramount concern. Arab countries must strengthen their social, economic, population and health policies to ensure greater gender equality in education, health and labour market outcomes. Moreover, financing to eliminate child marriage ensures human rights.