>>My name is Dr Zaki Wahhaj and I'm a senior lecturer in the School of Economics at the University of Kent. My research focuses on developing countries and in particular I'm interested in how social and cultural practices are sustained in societies and how they evolve. 

>>In today's lecture I'm going to talk about the practice of child marriage, which is prevalent in many parts of the world today. In a recent report from the United Nations Population Fund, it was reported that about 1/3 of women in developing countries around the world are married before the age of 18 and about one in nine of them are married before the age of 15. In particular countries the rates are significantly higher, for example in Niger it's 75 per cent, in Chad it's 72 per cent, in Bangladesh it's 66 per cent, and the rates are highest in sub-saharan Africa and in South Asia. The proportions are significantly lower for men than it is for women. Child marriage typically leads to early school dropout for girls, as well as early childbearing, drop out from work outside of the home and taking on domestic responsibilities within the household of the husband and the in-laws. The practice of child marriage is in fact illegal in many of the countries where it's prevalent but the laws are generally ignored and rarely enforced.

>>The Sustainable Development Goals of the ...of the United Nations which were declared in the UN summit last year actually has the elimination of child marriage as one of its targets within the broader goal of promoting gender equality. The reason that policymakers and governments are concerned about the practice of child marriage: first, it's a human rights issue because boys and girls who are made to marry before the age of 18 may not have reached the age and maturity to give informed consent; secondly, child marriage disrupts the
normal process of adolescence, or what economists call the disruption in the process of human capital accumulation.

>>Now, given these adverse consequences and the harmful effects of child marriage, we may ask why does the practice exist at all? Historically it has been more prevalent in certain parts of the world than in others: more prevalent in China, the Indian subcontinent, the Middle East and North Africa than, say, in Europe, and the sociological explanation for this is that in individualistic societies couples were expected to have their own source of income and the means of supporting a new household before they were expected to marry, whereas in collectivist societies young couples could expect the support of the extended family and their kinship group, and in fact they were encouraged to reproduce and had pressures to reproduce as the new children would strengthen the extended family and the kinship group, and this would lead to a practice of child marriage and early marriage. But, if this were the main reason for child marriage in these societies, then you would expect its prevalence to decline as these societies developed, as markets expanded and the family and kinship group became less important but in fact we see that the child marriage patterns in these societies are not evolving in the way that you'd expect given the rates of economic development. In a recent article with a co-author, we argue that the lives of women in patriarchal societies are actually ... that they seem to be constrained by multiple norms, such that the removal of one constraint, say, in the domain of education, would not necessarily lead to a change in practice in another domain such as the  practice of early marriage. 

>>And this is linked to a second sociological explanation for the prevalence of early marriage, which has to do with the idea that in patriarchal societies the honour of a family is linked to the purity of its women and brides, and in this case what purity means is that the women are virgins at the time of their marriage. 

>>Now, if this were the case, then it would lead to a practice of restricting contact between young girls and men outside of the family, and it would also lead to limiting their movement, as well as the practice of early marriage, and these pressures would survive in spite of economic development.

>>Now, as I said at the beginning of this talk, I'm an economist by training so how does the economic way of looking at social issues help us address the problem of child marriage? Well, the economic approach to social behaviour is to create models which focus on just the driving factors behind individual behaviour and then look at the aggregate effects of this behaviour and in the case of child marriage we can think about three economic ideas or factors which might influence behaviour. First is the idea of a marriage market, that is to say that there are many potential grooms who are seeking brides and many potential brides who are seeking grooms, and the idea of an equilibrium in this market. Second is the ... the idea of expectations about the future because parents who are deciding whether to marry off their daughter or to delay her marriage will take into consideration her future prospects on the marriage market, as well as the state of the marriage market in the future, and third is the idea of information asymmetry, which in this case captures a notion that certain information which is relevant for the marriage market, in particular whether the bride is a virgin or not, may be hidden or perceived to be hidden at the time that the match is being made, and this is going to have implications for the marriage market as well. Using these ideas we can create a simple model of the marriage market which will have the following driving mechanism: if potential brides have a certain characteristic which is not readily observed at the time that a match is made, then the prevalence of this characteristic is going to change with the amount of time that a potential bride has spent on the marriage market and in particular good qualities or desirable qualities are going to ... to decline with the age of the potential bride and this will create pressures for young girls to accept early offers of marriage rather than to postpone their marriage, and this can lead to a phenomenon of early marriage and the norm persisting over time in spite of changes in other domains. 

>>Now, to be able to use a model like this to analyse the effects of different policies, we need to have detailed data on actual marriage patterns in a particular society,in particular who is marrying whom, at what age are they marrying, and the terms of the marriage. During the last three years I have been involved in a project looking at the life choices of women in rural Bangladesh, in a project which is funded by Australian Aid, and as part of this project we collected information on the education, marriage patterns, work patterns of a large number of women, of 7,000 women in different parts of Bangladesh, which provides us the kind of data that we need to ... to put to use a kind of theoretical model along the lines that I described. From this data what we see is that, while the practice of child marriage is very prevalent in Bangladesh, it's not a static practice, it is actually changing over time. So in this graph what you can see is that the oldest women, who are in their late 30s, the prevalence of early marriage was actually very high, was above 80 per cent, but the youngest women in our survey, who are in their early 20s, the prevalence has declined significantly to just above 50 per cent and in these graphs we see that the marriage patterns are changing in other dimensions as well, so for example we see that the proportion of marriages which are arranged by a woman's parents or her family is declining, we see that marriage due to parental pressure has decreased and we're also seeing that the
difference in education, which used to be positive in favour of the husband, has actually been reversed and for the youngest women in our survey they actually have more education than their husbands, and if we look at the patterns of marriage before the age of 15, then we also see a shift all over the country. For the oldest women in our survey, we see that there were over 40 per cent of them who were married before the age of 15 in most parts ... in most parts of the country whereas that practice has declined quite significantly for the youngest women in our survey. 

>>The reasons why the marriage patterns in Bangladesh seem to have changed during the last 20 years, it's generally thought that it's because of a number of education-related programmes which were introduced by the government, as well as changes in the employment of women because of the rise in manufacturing and ...and so on. We can use this data to ... to choose the appropriate parameters for a theoretical model along the lines that I described and then … then the model can be used for ... for policy purposes to see what effect 
different kinds of policies related to education or other kinds of intervention will affect marriage market outcomes. 

>>So what can we learn from an exercise of this kind? I'm going to conclude with one key insight which is the following. Imagine that there is an intervention which improves access to schooling for girls or gives them employment training opportunities and that to be able to pursue these opportunities they have to postpone their marriage, they cannot pursue it once they're married. Now if such an intervention is available in one village or a small group of villages, then some girls might ... might take up the opportunity but not very many of them are going to do that because of the pressures to marry early that I described earlier. However, if the intervention is rolled out on a much larger scale, then many girls are faced with the same choices and if they all make the same kinds of decisions it's going to break the link between the age of a bride and her perceived quality that I described earlier, and that can lead to a shift in norms, a shift in the equilibrium in the marriage market, and lead to a much bigger change in the marriage patterns that we observe as compared to an intervention which took place in ... on a much smaller scale. 

>>I'm going to leave you with information regarding our website where you can find further information about this research project. Thank you very much.