Minimum marriage age laws and child marriage in Africa
Minimum marriage age laws, child marriage rates and adolescent birth:
Evidence from Sub-Saharan Africa
Abstract
This paper examines associations between minimum marriage age laws, child marriage rates and adolescent birth in 12 countries in sub-Saharan Africa (SSA). Two data sources are used: the Demographic and Health Surveys (DHS) from each country and the Child Marriage Database created by the Maternal and Child Health Equity (MACHEquity) research program at McGill University. The paper first assesses whether there is an association between consistent minimum marriage age laws of 18 years or older for girls and child marriage rates; and secondly, whether there is an association between consistent minimum marriage age laws of 18 years or older for girls and adolescent birth rates. Law consistency refers to where the general minimum marriage age, the minimum marriage age with parental consent and the age of sexual consent for girls, were all set at 18 years or older.Using multivariable models with robust standard errors to account for household clustering, we controlled for poverty, educational attainment, religion and rural or urban location. Holding these factors constant, the prevalence of child marriage and giving birth as an adolescent were 40 percent [prevalence ratio (PR) = 0.60, 95% confidence interval (95% CI) = 0.55 – 0.66] and 25 percent (PR= 0.75, 95% CI = 0.73 – 0.78) lower, respectively, in countries with consistent minimum marriage age laws of 18 years or over compared to countries with inconsistent laws.While still to be confirmed by quasi-experimental analyses, the results are consistent with the hypothesis that marriage laws protect against the exploitation of girls.
KEYWORDS: ADOLESCENT HEALTH, MARRIAGE POLICY, EARLY MARRIAGE, TEENAGE PREGNANCY, CHILDREN, ADOLESCENT MOTHER
1. Introduction
Child marriage is widely acknowledged as a harmful socio-cultural practice that is both a cause and outcome of human rights violations.1,2Defined as marriage or cohabitation before the age of 18 years old,3,4 child marriage undermines a girl’s ability to exercise her right to autonomy, live a life free from violence or coercion and it compromises her right to education.1,2,5 There is often explicit acknowledgement that the marriage is expected to result in childbirth.5,6 Thus child marriage also permits sexual exploitation under the guise of marriage and places a girl’s health at risk. Additionally, children of adolescent mothers start life at a disadvantage thus perpetuating a cycle of poverty and relative deprivation.5
Two main conventions aim to protect children in Sub-Saharan Africa (SSA), namely Article 1 of the 1989 United Nations Convention on the Rights of the Child (CRC) and Article 21(2) of the 1990 African Charter on the Rights and Welfare of the Child (ACRWC). All African countries except for Somalia have ratified the CRC, which describes several child rights violated by child marriage, without expressly referring to child marriage.7 African countries were however under-represented in the CRC drafting process and felt the need for a regional convention that complements the CRC, while addressing the specific realities of African children.8 The ACRWC thus emphasizes that signatories need to take effective action to end child marriage, set the minimum age of marriage at 18 years and make the registration of all marriages compulsory.9 As of January 2014, all African Union member countries had signed the ACRWC[a], though seven have yet to ratify it.9
The majority of SSA countries are, in principle, legally bound by the terms of these agreements. In practice, political will to change marriage laws varies considerably and efforts have been largely inconsistent. For example, 37 out of 41 SSA countries (90%) have nationally legislated minimum marriage ages of 18 years or older for girls.[b]However, 12 of these 37 countries allow marriage before age 18 with parental consent, allowing parents to marry off their daughters as a matter of course.[c] Furthermore, almost all SSA countries have exceptions allowing children to marry under customary law or other circumstances (such as pregnancy), without specifying an absolute minimum age for these marriages.
Inconsistent legal proscription is problematic because child marriage in Sub-Saharan Africa has been practiced for generations and is still seen as a culturally legitimate way of protecting girls.2,10In addition, factors contributing to the demand and supply of child brides continue to plague the region such as poverty, high fertility rates, poor educational opportunities, gender roles and economic shocks from unemployment or HIV/AIDS.6,10,11Consequently, poor parents with limited resources may have financial incentives for marrying girl children early, as the costs of raising and educating girls are weighed against the immediate promise of dowry. Allowing child marriage with parental consent and other exceptions may thus provide official sanctioning of a harmful custom. Setting clear and consistent laws against child marriage may be the first step in curbing the practice and possibly improving key population dynamics in sub-Saharan Africa.
A substantial and growing body of empirical evidence reveals clear associations between child marriage, adolescent birth and various maternal and child health outcomes. Girls who marry before the age of 18 years are more likely to have children as adolescents.10-14 Moreover, adolescent childbirth is associated with serious obstetric outcomes relative to adult childbearing.15-18 For instance, compared to women over the age of 20, girls giving birth between the ages of 10 and 14 years are five to seven times more likely to die from child birth, and girls giving birth between the ages of 15 and 19 years are twice as likely.19It bears repeating that the vast majority of adolescent births in sub-Saharan Africa occur within a marriage or union. Children of adolescent mothers also have poorer outcomes compared to older mothers, including lower birth weights, higher neonatal mortality rates, higher morbidity, stunting and mortality rates.16,20-23
Marriage before the age of 18 years is also an important determinant of women’s reproductive behaviour and health; compared to those who marry after age 18, child marriage is associated with a greater proportion of multiple unwanted pregnancies, higher total fertility rates, obstetric fistula, increased exposure to intimate partner violence and higher HIV prevalence rates.6,18,24-31Additionally, Clark et al.29 found that mean spousal age differences are higher among women who married as children relative to those who married as adults. This may constrain their ability to negotiate with their husbands and may also compromise control over their own reproductive health. In other words, child marriage adds ‘another layer of vulnerability’ over and above background characteristics associated with adolescent childbirth, such as lower education, poverty and living in rural areas.26
The literature reviewed shows that studies of child marriage mostly focus on South Asia resulting in little empirical evidence about sub-Saharan Africa. This is a significant oversight given the disproportionate burden of child marriage in the region.A United Nations Children’s Fund (UNICEF)report on child marriage found that 46% of 15–24-year-olds in South Asia were married before they reached 18, compared to 37% of 15 to 24-year-olds in sub-Saharan Africa.32However,15 of the 20 most affected countries are in sub-Saharan Africa (Table I). In Niger for example, three quarters of 20 to 24 year olds were married before the age of 18.32 In other words, sub-Saharan African countries have some of the highest rates of girl child marriage in the world but have been neglected from rigorous research on child marriage.
[Insert Table I]
There is global consensus on the importance of setting minimum marriage laws at 18 years, as well as substantial evidence of the link between child marriage, adolescent birth and reproductive health. Yet to the best of our knowledge, there are no studies that examine child marriage from a policy perspective. This study will thus make a much needed contribution to the literature by examining the association between minimum marriage age laws, child marriage and adolescent birth in Sub-Saharan Africa. The severe reproductive health and infant morbidity concerns exacerbated by child marriage and adolescent birth increase the need for analysis of prevention efforts, particularly those centered on marriage laws.
2. Sample
We merged information from two sources: the Demographic and Health Survey (DHS) and the Child Marriage Database. The DHS is funded by the United States Agency for International Development (USAID) in order to provide low- and middle-income countries with data needed to monitor and evaluate population, health and nutrition programs on a regular basis. This paper focused on responses to the DHS women’s questionnaire which collects information about socio-economic characteristics, reproductive history, maternity care, sexual activity and contraceptive use. Health histories and anthropometric data are also collected for children younger than five years of age.33
The longitudinal Child Marriage Database was created by McGill University’s MACHEquity[d]research program for 121 low- and middle-income countries currently included in the Demographic Household Surveys (DHS) and Multiple Indicator Cluster Surveys (MICS) by UNICEF.The Child Marriage Database, created through a systematic review of marriage legislation, civil codes and child protection legislation from 1995 to 2012, includes such indicators as: the legal minimum age for marriage as well as various exceptions permitting marriage at younger ages. Information is captured for both girls and boys, which allows for the analysis of gender inequalities in marriage laws. The study is limited to girls because child marriage affects more girls than boys and because many of the health risks are specific to girls, such as those arising from adolescent birth.5
The data source for the age of sexual consent is the Family Online Safety Institute’s (FOSI) Global Resource and Information Directory[e] (GRID). FOSI is an international non-profit focusing on comprehensive research of the issues, challenges and risks facing children online. GRID is an up-to-date source providing data and information on online safety, e-learning and country laws on sexual offenses, children and the use of the Internet in the commission of criminal activity.
Analyses were restricted to the most recent (2010-2012) DHS women’s datasets for sub-Saharan African countries, using 2009 child marriage policy data to better approximate a cross-sectional dataset. Additionally, the sample was restricted to women aged 15 years to 26 years as thisage cohort was more likely to have grown up in the global anti-child marriage era, signified by the 1989 Convention on the Rights of the Child (CRC) and Article 21(2) of the 1990 African Charter on the Rights and Welfare of the Child (ACRWC).
The countries and survey years examined were: Burkina Faso (2010), Burundi (2010), Cameroon (2011), Ethiopia (2011), Gabon (2012), Malawi (2010), Mozambique (2011), Rwanda (2010), Senegal (2010), Tanzania (2010), Uganda (2011) and Zimbabwe (2010).
2.1. Measures
The exposure is whethera country hadthree consistent minimum marriage age laws of 18 and over for girls; that is, where the general minimum marriage age, the minimum marriage age with parental consent and the age of sexual consent for girls, were all set at 18 years or older. The general minimum age refers to the age at which girls can get married without parental consent. The minimum age with parental consent refers to the age at which girls require their parents’ consent to marry. The age of sexual consent refers to the age at which a girl is legally capable of agreeing to sexual intercourse, so that an adult male who engages in sex with her cannot be prosecuted for statutory rape. Countries with all three laws set at 18 years or older were defined as having consistent laws against child marriage.
The outcome variables are child marriage and adolescent birth. Child marriage is a binary variable indicating whether age at first marriage or cohabitation was less than 18 years. Adolescent birth is a binary variable referring to whether age at first birth was less than 20 years. The primary question asked whether there is an association between living in a country withthree consistent minimum marriage age laws and the practice of child marriage. The secondary question asked whether there is an association between living in a country with three consistent laws and adolescent birth. A common set of controls were used:
•“Poor” is a series of indicator variables representing quintiles of household wealth, with the wealthiest quintile as the reference group; the wealth index, provided by the DHS, is based on ownership of specific assets (e.g., bicycle, radio and television), environmental conditions, and housing characteristics (e.g., type of water source, sanitation facilities, materials used for housing construction).
•Location is a binary variable defined as 1 if the household is located in a rural area and 0 if it is located in an urban area.
•Educational attainment is a binary variable defined as 1 if the highest level of school is secondary school or higher and 0 if it is primary school or less.
•Religion is a categorical variabledescribing whether the respondent is Christian, Muslim, not religious or follows traditional or animist religions. The reference category is Christianity. Religion is a country-specific variable in DHS with each country using different coding. The variable was harmonized by recoding all unspecified observations as missing.
•A binary variable for child marriagewas used in the secondary research question to identify respondents whose age at first marriage or cohabitation was less than 18 years.
2.2. Statistical Analyses
The study performeddescriptive analyses separately for each country as well as for all countries in the sample. Descriptive statistics include univariate statistics on the independent and control variables and bivariate associations between consistent minimum marriage age laws and the outcome variables. Multivariable log binomial regression models were used to examine the association between consistent laws and child marriage on the prevalence ratio scale. Log poisson regression models were used to estimate the association between law consistency and adolescent birth, because log binomial models failed to converge.34Independent variables were added one at a time in a series of models in order to test the strength of the coefficients; and results were presented as prevalence ratios (exponentiated regression coefficients) with 95% confidence intervals (CI) in parentheses.The highest variance inflation factor was 2.4 indicating little multicollinearity among thevariables.
Most DHS countries use a two-stage stratified cluster-sampling design to randomly select a fixed number of households for the survey, from primary sampling units that correspond to census enumeration areas. All eligible members in the household are asked to participate; that is, all women aged 15 to 49 years and men between 15 and 59 years.35This sampling technique may introduce intra-cluster effects, leading potentially to underestimating coefficient standard errors. For instance, women in a particular household or region may be more likely to have married before the age of 18 years than those in other households or regions.
There is no formal test to determine the appropriate clustering level for this type of survey design.36The general convention is to cluster at the highest level when possible in order to be conservative.37 We performed the analysis with robust variance clustered alternatively at the household and the country levels. The estimated standard errors were highly sensitive to this choice of clustering, with the country-level clustering being considerably less precise. In the primary analyses shown, the more precise household clustering likely underestimates the true uncertainty. The alternative choice of clustering at the country level, with relatively small numbers of exposed and unexposed clusters, likely overestimates the uncertainty.
Much of the published literature recommends the more conservative approach of overestimating uncertainty, based on the assumption that Type I errors are more costly than Type II errors in the context of statistical testing.36,37 We do not focus on statistical testing in this paper, and base interpretation more on the point estimates than on the associated intervals. While the point estimates may still be subject to endogeneity bias from unmeasured characteristics of countries with consistent laws against child marriage, the widening of the confidence intervals through the use of the robust clustering at the higher level does not correct this bias, but instead merely provides a more conservative interpretation of the estimates by representing them in the context of greater uncertainty. Faced with a choice between exaggerating the precision of our estimates and exaggerating the imprecision of our estimates, we opt for the less conservative depiction in our primary analyses because of our subjective beliefs that consistent laws are much more likely to be beneficial than harmful. Nonetheless, the estimated intervals in Tables VII and VIII should be viewed in light of this choice.
All descriptive analyses were weighted using the women’s individual sample weights and STATA version 12.Multivariate analyses were not weighted, so univariate and bivariate results are representative of the population, while multivariate analyses are only generalizable to the sample of women.38The final sample size in the 12 countries surveyed is 79 567 women aged between 15 and 26 years old. We restricted our analyses to women with non-missing information on key covariates (n=78 951). Analyses involving child marriageand adolescent birth were performed on the entire sample of women; and those involving religion were largely restricted to eleven countries (74 174 women), as Tanzania does not capture religious information in their DHS surveys.