Vous consultezToo Poor to Marry?

AuteurAnne-Emmanuèle Calvès[*] [*] Sociology Department, University of Montreal
Translated...
suitedu même auteur

1 - Worsening job prospects in cities

5 Burkina Faso is one of the world’s poorest countries and its economy has declined over the last fifteen years. Although “one cannot really speak of a collapse in the economy” (Diabré, 1998, p. 29) as in some other countries of sub-Saharan Africa, a number of sectors were clearly in difficulty as early as the late 1980s (Chambas et al., 1999). To cope with this deterioration, in 1991 the country adopted a structural adjustment programme that included the restructuring and privatization of semi-public enterprises, reform of the public sector and devaluation of the CFA franc in 1994. This restrictive economic programme, combined with the slowdown in the economy and the urban population increase profoundly altered the urban labour market. The reform of the public sector froze wages and cut back recruitments (Gaufryau and Maldonado, 2001). Restructuring and privatization also led to business closures and lay-offs in the semi-public sector (Diabré, 1998; Sanou, 1993). Although it is hard to evaluate, the share of the informal sector in the country’s economy appears to have grown in recent years. According to Charmes (1996), 77% of jobs, excluding farm work, were in the informal sector in Burkina Faso in 1990, compared with 70% in 1980. In Ouagadougou, over 80% of paid jobs were outside the formal sector in the early 1990s (Sananikone, 1996). Apart from this increase in the informal sector’s share of the economy, the range and proportion of insecure jobs in the informal sector is reported to have grown in Burkina Faso as in other sub-Saharan countries (Charmes, 1996; Meagher, 1995). Small urban businesses in the informal sector appear to have become less viable since devaluation (Camilleri, 1997).

6 As in the region’s other countries (Marcoux and Piché, 1998; Antoine et al., 2001; Kouamé et al., 2001), younger cohorts have been particularly affected by this deterioration in Burkina Faso’s urban labour market. One of the most obvious signs of this is their high unemployment rate. According to a recent study, unemployment among young city-dwellers in Burkina Faso has increased significantly in the last ten years; nearly 40% of men and 55% of women aged 15-24 not enrolled in education in 2000 were excluded from the paid labour market (Calvès and Schoumaker, 2004). In addition to the shrinking job market, the economic crisis has also seriously affected the quality of jobs that young people can take. Since the formal sector cannot generate enough employment to absorb all young people arriving on the labour market, younger cohorts, especially the qualified, are increasingly turning to the informal sector (Calvès and Schoumaker, 2004). Even if earnings in this sector are not always lower than in the formal sector and working in the informal economy is not always synonymous with poverty (Charmes, 2000), young people do tend to be over-represented in the least profitable and most vulnerable sectors of the informal economy. Informal activities, other than crafts and retail, such as bicycle and car repair for young men and hairdressing and domestic help for young women, attract growing numbers of young people (Charmes, 1994; Calvès and Schoumaker, 2004).

7 How has this deterioration in employment conditions since the early 1990s affected the union formation of young city-dwellers? A number of studies have shown that the age at first union for Burkinabè women has clearly risen in the last twenty years, particularly among women living in towns (Tabutin and Schoumaker, 2004; Calvès et al., 2007). For men, the data from national Demographic and Health surveys (DHS) (INSD and Macro International, 2000; INSD and ORC Macro, 2004) show that median age at first union in urban areas rose from 26.6 to 29.8 between 1998 and 2003. But little is known about the role of urban employment changes in this increase. Individual-level data on employment and marital status in sub-Saharan Africa are often collected separately and are hard to compare. The few analyses of event history survey data that contain both marital and employment trajectories for individuals have, however, shown the determining effect of economic activity on union formation among younger cohorts of African city-dwellers (Antoine et al., 1995; Marcoux and Piché, 1998; Lardoux, 2004). In Dakar, for example, these analyses show that it is primarily economic factors, especially unemployment, that delay marriage for the younger male cohorts. In Burkina Faso, although the relation between worsening employment conditions and rising age at marriage has never been the subject of quantitative analysis, the qualitative data suggest that many young city-dwellers are forced to live with their parents longer and to remain single, because “they can’t afford” to do otherwise. On the basis of interviews with young single men in the popular districts of Ouagadougou, Sévédé-Bardem (1997) describes the frustration many of them feel with respect to marriage and relations with the opposite sex in general. Disappointed and demoralized by the insecurity of their economic activities, a number complained that “girls these days marry guys who have money, otherwise they don’t love them (p. 32)”.

2 - Data and methodology

8 To analyse the changes in union formation among young city-dwellers in Burkina Faso, the present study uses data from a national retrospective survey on migration dynamics, urban integration and the environment in Burkina Faso (Dynamiques migratoires, insertion urbaine et environnement au Burkina Faso, EMIUB) carried out with 8,644 respondents in 2000[1] [1] This survey is part of a wider research project initiated...
suite. The analysis presented here focuses on urban areas, and only respondents living in the country’s two largest cities, Ouagadougou and Bobo-Dioulasso, were included. With populations estimated at 1,066,100 and 410,500 respectively in 2006, these two cities alone account for 70% of the country’s urban population and 11% of its total population in 2006. In addition to the size of their populations, the maturity, structure and profile of their socio-economic activities and the level of access to services give these cities an undeniable urban status, setting them apart from other urban centres that may fall into the categories of semi-urban village or small town depending on the definition adopted (Beauchemin et al., 2002).

9 Although the survey, as its name suggests, was primarily concerned with migration and urban integration, the data collected can be used to analyse changes in men’s entry into first union and the effect on this process of changes in employment, since they include full retrospective marital and employment histories and other biographical information such as residential history for all respondents. More specifically, in the marital section of the survey, each respondent aged 15-64 was asked about all the unions they had formed, from the first to their current marital status. In Burkina Faso, as in most sub-Saharan countries, entering a union is a process rather than a single event (Meekers, 1992). A number of stages and ceremonies usually mark the transition from being single to the socially recognized status of being “married”. Among the Mossi of Burkina Faso, for example, the process traditionally begins with engagement, symbolized by a visit from the suitor, who brings presents and gifts of money to the head of the girl’s family. If her father agrees to the marriage, bridewealth, often paid in instalments, is offered to the bride’s family and the union is sealed by a traditional and/or religious wedding after which the couple may begin to cohabit (Pageard, 1969). In a number of African countries, especially in cities, a civil wedding ceremony is sometimes added to the traditional and religious ones, which are still major celebrations that often take place before the legal wedding (Antoine, 2002). The existence of these various celebrations, which may be simultaneous or spread out over time, makes it hard to establish the starting date for a marriage. Unlike the Demographic and Health Surveys, which often merely record age at first marriage as reported by the respondents, the EMIUB survey collected information dating each stage of union formation: first cohabitation, religious, customary or civil weddings as the case may be. For the purposes of this article, entry into a union is marked by one or more of these events.

10 The questionnaire section covering the working career is also highly detailed and collects information about all periods of education, apprenticeship, economic activity (paid or unpaid jobs) and unemployment of more than three months since the respondent’s sixth birthday. For each period of activity, further information was collected about the type of activity (open-ended question), whether respondents were paid for the work done and whether they received or issued payslips.

11 The present analysis comprises three parts: the first concerns changes in the timing and procedures of entry into first union (first matrimonial stage) for men in urban areas; the second examines changes in the timing of all the processes in union formation (second and third stages); and the third addresses the effect of men’s employment status on the formation of their first union and how this effect varies across cohorts. To analyse changes in the timing and nature of the first union, the experience of young male city-dwellers today, namely respondents aged 25-34 at the date of the survey (born between 1965 and 1974), is compared with that of previous cohorts at the same age. Each cohort in the total sample of 958 men comprises male respondents living in the two cities aged 25-34. Since this was a national retrospective survey, it covered internal migration and included in the sample respondents who lived in the cities between the ages of 25 and 34, even if they had moved away after that age. Respondents born between 1936 and 1944 were excluded from the analysis because data collected from the oldest cohorts are more subject to omission and inaccurate dating of events. Respondents of the 1965-1974 cohort were observed until 2000, those of the 1955-1964 cohort until 1990 and those of the 1945-1954 cohort until 1980. To analyse the changes in the timing of the process of first union formation, the marital status of the three cohorts of men is compared at three points in time (1980, 1990 and 2000) and survival curves are calculated using the Kaplan-Meier estimator of survival function (Courgeau and Lelièvre, 1989). Also examined are the way the process is initiated (type and number of matrimonial events) and the intervals between them.

12 The final part of the analysis concerns the determinants of the hazard of forming a union, and particularly the effect of employment status and type of work, using Cox semi-parametric proportional hazards models (Courgeau and Lelièvre, 1989; Allison, 1984). In these regression models, the dependent variable is the hazard rate for a first union (cohabitation or wedding) and the main independent variable is employment status. Using the full occupational history, an independent variable was used whose value may vary over time and which reflects the successive jobs of young respondents during the observation period (from 15 to first union or survey date). It combines both young men’s employment status (paid or unpaid) and their employment sector. An individual may, for example, be first exposed to the hazard of forming a first union as a person “without paid employment” and then be classified some years later as “craft worker”.

13 Although there is no single definition of the informal sector in urban Africa, it is generally agreed that the activities of the informal economy sidestep government fiscal and welfare contribution regulations (Dijkman and Van Dijk, 1993; Osirim, 1992). Most of the time, the independent workers, employees and small firms of the informal sector are not registered. For practical reasons, public sector employees and all workers who receive payslips are considered in this study to be in the formal sector. The employers who issue payslips to their employees are also included in this sector. The formal sector is then divided into public and private. Given that most workers are to be found in the informal sector, and that the degree of differentiation within this sector appears to be increasing (Charmes, 1996), a more detailed classification of informal urban activities was required. They are therefore classified in four categories: craft workers (tailor, carpenter, shoemaker, brickmaker, etc.), small-scale retail (sale of clothes, water, miscellaneous small items, etc.), farming, and services (bicycle and car repair, driver, etc.).

14 On the basis of the few studies that have examined the determinants of first marriage among men in sub-Saharan Africa (Antoine et al., 1995; Marcoux and Piché, 1998; Lardoux, 2004), a number of socio-cultural factors likely to influence first union formation were also integrated in the multivariate analysis as control variables. For example, Antoine et al. (1995) have shown that in Dakar, Muslim men marry earlier than Christians and that there are also differences between ethnic groups. Although the effects of these cultural variables on men’s marriage are not significant in Bamako (Lardoux, 2004; Marcoux and Piché, 1998), in both Dakar and Bamako the level of education has a marked, significant effect on men’s first entry into union. The three studies cited above agree that men with more years of schooling marry significantly later than the others.

15 The control variables selected for the multivariate model were consequently religion (Christian or other), ethnic group (Mossi or other) and educational level (none or Koranic school, primary, secondary or higher). In the light of the highly marked differences between the marriage patterns of men according to their place of residence (Lardoux, 2004), place of residence (urban or rural) during childhood (age 6) was also included as a control variable for the model[2] [2] Since the EMIUB survey did not collect data on men’s birth...
suite. Unlike employment status, these independent variables are fixed and reflect individuals’ characteristics at the start of the observation period (age 15). Finally, in addition to the general model including men of all cohorts, separate models were estimated for each cohort to evaluate the differential effect of employment on first union formation over time.

16 Table 1 presents the socioeconomic characteristics observed for the independent variables included in the multivariate models. They correspond to the characteristics of the men in each cohort when aged 25-34 (in 2000, 1990 and 1980). Thanks to the development of education in Burkina Faso since the late 1960s[3] [3] The primary school enrolment rate rose from 13% in 1970...
suite, men in the latest cohort have a notably higher level of education than in the earlier ones. Whereas a large majority (67%) of individuals in the 1965-1974 cohort attended modern schools at least until the end of primary school, this was the case for only 51% of the 1955-1964 cohort and 39% of the 1945-54 cohort. In confirmation of the findings in earlier studies (Calvès and Schoumaker, 2004; Charmes, 1996), the table shows that although they are more educated, young men in the 1965-1974 group have found it more difficult than the earlier groups to enter the labour market. Among city-dwellers aged 25-34 in 2000, more (18.5%) had no paid job than among the earlier cohorts at the same age (9%). When they did have a job, fewer worked in the formal sector of the economy than before: 22% of men in the 1965-1974 cohort had a job in the formal sector, compared with 36% of the 1955-1964 cohort and 37% of the 1945-1954 cohort. Note that this decline in formal employment mainly concerns the public sector, since the proportion of jobs held by young men in the private sector remained relatively stable from 1980 to 2000. A large number of respondents (nearly 70% on average) grew up in rural areas, which is not surprising considering the major role of internal migration in Burkina Faso’s urbanization process. As to be expected, the proportion of men who grew up in cities is highest (over one-third) in the youngest group. Religious and ethnic features do not vary significantly between the groups and the respondents are mostly Mossi (66%) and Muslim (60%)[4] [4] The “Other” category of the religion variable (which...
suite.

Table 1 - Socioeconomic characteristics of men aged 25-34 by cohort (weighted data, percentages)

3 - First union occurs later and in a different manner

17 Figure 1 shows the marital status of men in each cohort when they were aged 25-34 (in 2000, 1990 and 1980). City-dwellers were less likely to marry before the age of 35 in 2000 than ten or twenty years earlier. Only 44% of men aged 25-34 in 2000 were married compared with 57% in the 1955-1964 cohort and 70% in the 1945-1954 cohort at that age. Although men are much more likely to be unmarried, cohabitation, as in other African cities (Thiriat, 1999), is also increasing in the Burkina Faso cities: 14% of men in the most recent cohort reported that they were cohabiting in 2000 compared with only 4% of the earliest cohort.

Marital status of men aged 25-34 by cohort (1980-2000), weighted data

Marital status of men aged 25-34 by cohort (1980-2000), weighted data

18 The Kaplan-Meier curves showing the proportion of the cohort at each age that has not yet formed a first union (as marked by one of the three wedding ceremonies or cohabitation) confirm the rise in age at first union by cohort (Figure 2). Before age 20, the hazard of entry into union is roughly the same for the three cohorts, then the gap between them rapidly widens in a statistically significant manner after that age. Whereas 71% of the most recent cohort were still unmarried at age 25, this was only true of 65% of the middle cohort and 61% of the earliest cohort. At age 30, the gap is even more marked, with 40.3% of men in the 1965-1974 cohort still single compared with 28.5% and 20.2% for the earlier cohorts. The median age at first union (when half are married) has risen from 26.2 years for the earliest cohort to 28.2 for the most recent.

Men’s first union: Kaplan-Meier survival curves, by cohort

Men’s first union: Kaplan-Meier survival curves, by cohort

19 Not only do young city-dwellers in Burkina Faso form their first union significantly later than in the past, but they also begin that union differently (Table 2). In confirmation of the results in Figure 1, city-dwellers in the latest cohort are much more likely than the earlier ones to begin their union by a period of cohabitation. One-third of men in the 1965-1974 cohort began their first union by cohabitation without a wedding, compared with 18% of the 1955-1964 cohort and only 7% of the 1945-1954 cohort.

Table 2 - Type of celebration of men’s first union by cohort (weighted data in %)

20 In addition to the increase in cohabitation (or periods of cohabitation), the survey data show that young city-dwellers are less likely than in the past to choose (or “be able to afford”) to mark their first union by two or three wedding ceremonies within one month. Whereas one-third of young men in the 1945-1954 cohort (34%) did so, for the middle cohort it was only 27%, and for the youngest cohort only 18% married in this way.

21 Just like their predecessors, however, young city-dwellers who do marry go for traditional and religious weddings. In all cohorts the most frequent marriage ceremony is the traditional wedding, whether celebrated in the same month as or before the religious wedding. In the youngest cohort, more men also postpone the traditional wedding and begin their marriage with a religious wedding (21%) than in the past (only 15% in the oldest cohort). Civil ceremonies, rarely practised by the earlier cohorts, attract less than 5% of the most recent cohort.

4 -  A more extended process of union formation

22 After analysing the event that begins the process of union formation, the Kaplan-Meier estimator was used to examine changes in the timing of the successive events (cohabitation, religious, traditional or civil wedding). Here the object of study is the second stage in the process (a second event of whatever sort) among men who have already passed the first stage, and the third stage among those who have already passed two stages. Table 3 shows the proportion of men in each cohort who have not yet gone through a second matrimonial event one, three and six months after the first, and the proportion of men in each cohort who have not yet gone through a third event one, three and six months after the second.

Table 3 - Proportions of men who have not passed various stages in the matrimonial process over time (Kaplan-Meier estimators)

23 Men in the most recent cohort form unions later than their predecessors and the process of forming a union is significantly longer (Table 3). Six months after the start of their union, only 19% of the men in the earliest cohort had not moved on to a second matrimonial event, whereas this is the case with 46% of the youngest. The third stage in the process is also postponed: of those who had passed two stages, one-quarter of the youngest had not passed the third stage six months after the second, compared with only 9% of those in the 1955-1964 cohort and 6% of the 1945-1954 cohort.

5 - Effect of employment on first union formation

24 Table 4 shows the results of the analysis of the hazard of entering a first union. Four Cox semi-parametric models were estimated: the first evaluates the effect of employment on the hazard of first union formation, all other things being equal, for all the men in the sample, and the other three give the results of the same analysis for each of the cohorts separately[5] [5] The overall model was estimated separately for men living...
suite. The Table shows the exponential regression. A variable that increases the hazard of forming a first union multiplies the basic hazard by a coefficient greater than 1; a variable that reduces the hazard has a coefficient of less than 1.

Table 4 - Factors affecting men’s first union formation (Cox semi-parametric proportional hazards models, exponential, weighted data)

25 As in Dakar (Antoine et al., 1995) and Bamako (Marcoux and Piché, 1998; Lardoux, 2004), having a paid job significantly advances the process of forming a first union among men in Ouagadougou and Bobo-Dioulasso (Table 4). However, the paid job effect on the hazard of forming a first union varies with the type of employment. Workers in the formal sector appear to have particular advantages in the marriage market. Across all cohorts, a public service job multiplies the hazard of forming a first union by 2.6 and a job in the private formal sector multiplies it by 1.7. Small traders in the informal sector also start their life as a couple earlier than men with no paid job. However, a paid job is not always a factor advancing the process of first union formation, and the presence of a hierarchy among informal sector jobs is confirmed by these figures. Controlling for individual characteristics (cohort, education, religion, ethnic group, childhood residence), the hazard of forming a first union for workers in the informal farm sector and informal service sector is not significantly different from that of men with no paid job.

26 Among the control variables, only education significantly affects early first union formation. In confirmation of earlier studies on the determinants of city-dwellers’ first union formation (Antoine et al., 1995; Marcoux and Piché, 1998; Lardoux, 2004), cultural variables in Ouagadougou and Bobo-Dioulasso have little effect on men’s marriage but post-primary education does significantly delay their first union. Men who have never attended a modern school marry 1.7 times earlier than those with secondary or higher education. Although the observed probability of contracting a first union varies significantly between cohorts (Figure 1), the cohort effect disappears when fixed socioeconomic characteristics (mainly education) and employment status for all men are considered. The later formation of a first union among men in the most recent cohort is therefore due to their longer school enrolment but also their less stable employment status.

27 To evaluate more closely the importance of having a job in the formation of couples by young city-dwellers, separate models were estimated for each cohort. As Table 4 shows, for men born between 1945 and 1954, type of work and employment status do not significantly affect the likelihood of forming a first union. Indeed no discriminating factor is revealed in the analysis of first unions in this city-dwelling cohort. All the men in this cohort formed their first union at about the same age and seem to have been relatively “equal” in the marriage market. As in Dakar (Antoine et al., 1995), the formation of a first union among men in the earlier cohorts in Ouagadougou and Bobo-Dioulasso appears to be linked more to their life-cycle (old enough to marry) than to financial security.

28 The same holds for men in the middle cohort (1955-1964). For most of them, having a paid job is not a major asset in beginning the matrimonial process, except for civil servants, who form a union 2.5 times earlier than the unemployed. Post-primary education, on the other hand, delays union formation.

29 As can be seen in the last model in Table 4, the role of employment in men’s first union formation has clearly changed in Ouagadougou and Bobo-Dioulasso in recent years. Whereas employment had little or no effect on first union formation among their predecessors, having a paid job of any sort facilitates in a statistically significant manner the beginning of the matrimonial process for young men in the most recent cohorts. Although young civil servants have the greatest advantage in the marriage market (with entry into union 3.3 times faster than the unemployed), informal jobs in crafts, small trade and even farming advance the formation of a first union among younger city-dwellers quite as much as being an employee in the private sector (hazard twice as high as for the unemployed). Jobs in services, the least profitable part of the informal sector, are also the least advantageous in the marriage market, although having a job there rather than none at all multiplies by 1.6 the chance of forming a couple for the youngest men. Just as for those in the previous cohort (1955-1964), the men in the most recent cohort who have never attended school or only Koranic school form a union 1.7 times faster than those who stayed in education until secondary or higher level.

Conclusion

30 Analysis of the biographical data in the EMIUB survey reveals major quantitative and qualitative changes in the process of first union formation among city-dwellers in Burkina Faso. Men in Burkinabè cities enter a union much later than before and their manner of doing so has also altered: periods of cohabitation are more frequent and some ceremonies are postponed or omitted by the latest cohort of bridegrooms. The timing of the second and third matrimonial stages has also changed and the entire process is tending to take longer.

31 This later first union formation among men in the most recent cohort can be partly explained by the deterioration of their employment position. Multivariate analysis by cohort also shows that whereas for the earliest cohort the date of first marriage was not a matter of money, for the most recent, obtaining a paid job is crucial to forming a first union. The growing individualization of financial responsibility for wedding costs observed in many African countries is probably one factor behind the greater importance of employment in forming a first union. A number of studies have noted that the cost of weddings, including the various ceremonies and payment of the bridewealth, is increasingly borne by the groom alone rather than by his family (Lardoux, 2005; Adjamagbo and Delaunay, 1999; Isiugo-Abanihe, 1994). Not only are wedding costs becoming more individual, they also appear to be rising in a number of African countries (Lardoux, 2005; Isiugo-Abanihe, 1994), including Burkina Faso (Sévédé-Bardem, 1997). Given this background, it is hardly surprising that obtaining a paid job has become a necessary condition for forming a first union among men in the most recent cohorts. Echoing the frustration of young bachelors in the popular districts of Ouagadougou expressed in Sévédé-Bardem’s study (1997), our results also show that, with increasing unemployment and informalization in urban areas, inequalities with respect to marriage are emerging among men in the most recent cohorts according to their employment status and type of work they do. Forming a first union has become particularly difficult for young men who are unemployed or hold unstable jobs in the informal sector. The changes observed in the cost of weddings and the people who pay for them probably also contribute to these inequalities in the marriage market. As the increasing cost of weddings is borne more by the individual, having a paid job is crucial, though the level of income generated by this job also affects young city-dwellers’ ability to save the money needed to form a union.

32 These results suggest a number of avenues for future research on first union formation among men in Burkina Faso. The changing manner in which first unions are formed, as shown here, raises a number of questions. What is the social status of unions that consist of cohabitation without wedding ceremonies, and what situations do they cover? As certain wedding ceremonies are postponed and spread out over time, is the institution of marriage becoming less important?

33 The determinants of delayed first marriage also need further research. Although it appears clear that the late marriages of the youngest cohorts are partly “forced” upon many men who cannot afford to marry, it is worth examining whether for other men this is perhaps a choice that reflects new aspirations and a new conception of marriage and the couple. It would be particularly useful to have further data on the motivations behind decisions to form a union and the experience of single life among young men in various socio-occupational categories. Given that procreation remains the prime purpose of marriage, it is worth examining the consequences of these changes in the marriage process on men’s entry into fatherhood and family life. Earlier studies have shown that longer single life and more informal first unions have given young men greater room for manœuvre in recognising and financially supporting their children born outside marriage (Launay, 1995; Calvès, 2000). More generally, we need to study how the changing matrimonial status at first birth in Burkinabè cities will affect paternity, relations within the couple and the links between young fathers and their children born outside marriage.

34 * * *

Bibliographie

References

Adjamagbo Agnès, Valérie Delaunay, 1999, “Une approche qualitative de l’évolution des modèles familiaux dans une population rurale sénégalaise”, Document de recherche, 6, Équipe de recherche Transition de la fécondité et santé de la reproduction, Paris, Institut de recherche pour le développement.

Allison Paul D., 1984, Event History Analysis: Regression for Longitudinal Event Data, Beverly Hills, California, Sage Publication.

Antoine Philippe, 2002, “L’approche biographique de la nuptialité : application à l’Afrique”, in G. Caselli, J. Vallin and G. Wunsch (eds.), Démographie : analyse et synthèse (Volume II : Les déterminants de la fécondité) Paris, INED.

Antoine Philippe, Mamadou Djiré, Benoît Laplante, 1995, “Les déterminants socio-économiques de la sortie du célibat à Dakar”, Population, 50 (1), pp. 95-117.

Antoine Philippe, Mireille Razafindrakoto, François Roubaud, 2001, “Contraints de rester jeunes ? Évolution de l’insertion dans trois capitales africaines : Dakar, Yaoundé, Antananarivo”, Autrepart, 18, pp. 17-36.

Beauchemin Cris, Estelle Beauchemin, Gaële Le Jeune, 2002, “Tabvilles-BF : Rapport de présentation”, EMIUB, Document technique d’analyse, 2002-1, Université de Montréal, Département de démographie, 19 p

Calvès Anne-Emmanuèle, 2000, “Premarital childbearing in urban Cameroon: paternal recognition, child care and financial support”, Journal of Comparative Family Studies, 31 (4), pp. 443-452.

Calvès Anne-Emmanuèle, Jean-François Kobiané, Edith Martel, 2007, “Changing transition to adulthood in urban Burkina Faso”, Journal of Comparative Family Studies, 38 (2), pp. 265-283.

Calvès Anne-Emmanuèle, Bruno Schoumaker, 2004, “Deteriorating economic context and changing patterns of youth employment in urban Burkina Faso: 1980-2000”, World Development, 32 (8), pp. 1341-1354.

Camilleri Jean-Luc, 1997, “L’impact de la dévaluation sur la petite entreprise au Burkina Faso”, Note de recherche n° 97-62, Paris, Réseau Entrepreneuriat de l’AUF.

Chambas Gérard, Jean-Louis Combes, Patrick Guillaumont, Sylviane Guillaumont, 1999, Burkina Faso : les facteurs de croissance à long terme, Emerging Africa research programme, Paris, OECD.

Charmes Jacques, 1994, “Les jeunes et l’économie au Sahel : place, rôle et attentes”, Pop Sahel, 21, pp. 30-35.

Charmes Jacques, 1996, “Emploi, informalisation, marginalisation ? L’Afrique dans la crise et sous l’ajustement, 1975-1995”, in J. Coussy and J. Vallin (eds.), Crise et population en Afrique, Paris, Ceped, pp. 495-519.

Charmes Jacques, 2000, Informal sector, poverty and gender, a review of the empirical evidence, http://www.wiego.org/papers/charmes3.pdf.

Courgeau Daniel, Éva Lelièvre, 1989, Analyse démographique des biographies, Paris, Éditions de l’Ined, 270 p.

Diabré Zéphirin, 1998, “The political economy of adjustment in Burkina Faso”, CAER II, Discussion Paper 28, Harvard Institute for International Development.

Dijkman Hanneke, Meine Pieter Van Dijk, 1993, “Female entrepreneurs in the informal sector of Ouagadougou”, Development Policy Review, 11 (3), pp. 273-88.

Gaufryau Bertrand, Carlos Maldonado, 2001, “Burkina Faso”, in B. Gaufryau, C. Maldonado (eds.), L’économie informelle en Afrique francophone : structure, dynamiques et politiques, Geneva, International Labour Office, pp. 147-183.

Hertrich Véronique, 1998, “Are men’s and women’s answers to be equally trusted? A dual collection of birth and marriage histories in a population in Mali”, Population, An English Selection 10 (2), pp. 303-318.

Insd, 2004, Projection de la population au Burkina Faso, Ouagadougou, Institut national de statistique et de démographie.

Insd and Macro International Inc., 2000, Enquête démographique et de Santé, Burkina Faso 1998-1999, Calverton, Maryland, (USA), Macro International Inc.

Insd and Orc Macro, 2004, Enquête démographique et de santé du Burkina Faso 2003, Calverton, Maryland (USA), INSD and ORC Macro.

Isiugo-Abanihe U.C., 1994, “Consequences of bridewealth changes on nuptiality patterns among the Ibo of Nigeria”, in C. Bledsoe and G. Pison (eds.), Nuptiality in Sub-Saharan Africa: Contemporary Anthropological and Demographic Perspectives, Oxford, Clarendon Press, pp. 74-91.

Kouamé Aka, Ngoy Kishimba, Mathias Kuépié, Donatien Tameko, 2001, “Crise, réformes des politiques économiques et emploi à Yaoundé”, Les dossiers du CEPED, 64, Paris, Ceped.

Lardoux Solène, 2004, “Marital changes and fertility differences of women and men in urban and rural Mali”, African Population Studies, 19 (2), pp. 89-123.

Lardoux Solène, 2005, “Marriage and co-residence in Muslim West Africa: A Malian case study”, paper presented at the 25th International Population Conference, Tours, France, 18-23 July.

Launay R., 1995, “The power of names: illegitimacy in a Muslim community in Côte d’Ivoire», in S. Greenhalgh (ed.), Situating fertility, Anthropology and Demographic Inquiry, Cambridge University Press, pp. 108-129.

Marcoux Richard, M. Gueye, M.K. Konaté, 1995, “La nuptialité : entrée en union et types de célébrations à Bamako”, in D. Ouédraogo and V. Piché (ed.), L’insertion urbaine à Bamako, Paris, Éditions Karthala, pp. 117-144.

Marcoux Richard, Victor Piché, 1998, “Crise, pauvreté et nuptialité à Bamako (Mali)”, in F. Gendreau (ed.) Crises, pauvreté et changements démographiques dans les pays du Sud, AUPELF-UREF-Editions ESTEM, pp. 219-236.

Meagher Kate, 1995, “Crisis, informalization and the urban informal sector in sub-Saharan Africa”, Development and Change, 26 (2), pp. 259-284.

Meekers Dominique, 1992, “The process of marriage in African societies: A multiple indicator approach”, Population and Development Review, 18, pp. 61-78.

Meekers Dominique, Anne-Emmanuèle Calvès, 1997, “Main girlfriends, girlfriends, marriage, and money: the social context of HIV risk behaviour in sub-Saharan Africa”, Health Transition Review, supplement to volume 7, pp. 361-376.

Nrc, 2005, Growing up Global. The Changing Transitions to Adulthood in Developing Countries, C.B. Lloyd (ed.), Committee on Population and Board on Children, Youth and Families, Division of Behavioural and Social Sciences and Education, Washington D.C., The National Academies Press, 700 p.

Osirim Mary J., 1992, “The state of women in the Third World: The informal sector and development in Africa and the Caribbean», Social Development Issues, 14 (2/3), pp. 74-87.

Pageard Robert, 1969, “Le droit privé des Mossi : tradition et évolution”, Recherches voltaïques, no. 11, Paris, CNRS and Ouagadougou, CVRS.

Poirier Jean, B. Dabiré, Gaëlle Le Jeune, 2001, “Projet d’étude des stratégies de reproduction des populations sahéliennes à partir de l’enquête Dynamique migratoire, insertion urbaine et environnement au Burkina Faso”, Cahiers québécois de démographie, 30 (2), pp. 289-309.

Sananikone O, 1996, “Burkina Faso”, in L. Webster, P. Fidler (ed.), The Informal Sector and Micro-Finance Institutions in West Africa, Washington D.C., World Bank, pp. 81-90.

Sanou Ourobé Mathias, 1993, Pauvreté et marché du travail à Ouagadougou (Burkina Faso), Labour Institutions and Development programme, Geneva, International Institute for Labour Studies.

Sévédé-Bardem Isabelle, 1997, Précarités juvéniles en milieu urbain africain (Ouagadougou), Paris, L’Harmattan.

Silberschmidt M., 1992, “Have men become the weaker sex ? Changing life situations in Kisii District, Kenya”, The Journal of Modern African Studies, 30 (2), pp. 237-253.

Tabutin Dominique, Bruno Schoumaker, 2004, “The demography of sub-Saharan Africa from the 1950s to the 2000s. A survey of changes and a statistical assessment”, Population-E, 59 (3-4), pp. 457-556.

Thiriat Marie-Paule, 1999, “Les unions libres en Afrique sub-saharienne”, Cahiers québécois de démographie, 28 (1/2), pp. 81-115.

World Bank, 1994, World Development Report 1994: Infrastructure for Development, Washington D.C., Oxford University Press.

Notes

[ *] Sociology Department, University of Montreal
Translated by Roger Depledge

[ 1] This survey is part of a wider research project initiated by the Demography Department at the University of Montreal in collaboration with the Unité d’enseignement et de recherche en démographie, (demography teaching and research unit, UERD) at the university of Ouagadougou and the Centre de recherche sur la population et le développement, (population and development research centre, CERPOD) in Bamako. For more on the survey, see Poirier et al. (2001).

[ 2] Since the EMIUB survey did not collect data on men’s birth history, it is not possible to include the presence of children born outside marriage as an independent variable in the multivariate model.

[ 3] The primary school enrolment rate rose from 13% in 1970 to 30% in 1992 (World Bank, 1994).

[ 4] The “Other” category of the religion variable (which mainly includes animists) accounts for less than 1% of respondents: it has been included in the “Muslim” category in the multivariate models, which thus distinguishes between Christians and non-Christians.

[ 5] The overall model was estimated separately for men living in Ouagadougou and Bobo-Dioulasso and the results are similar for the two cities. Given the small number of respondents living in Bobo-Dioulasso (80% of the sample were Ouagadougou residents), it was not, however, possible to do separate cohort analyses for the two cities.

Résumé

En s’appuyant sur une enquête biographique détaillée réalisée en 2000 au Burkina Faso (EMIUB), cet article étudie l’effet de la détérioration des conditions d’emploi en milieu urbain au cours des années 1990 sur l’entrée en première union des hommes. Les résultats montrent que la première mise en couple des jeunes citadins est aujourd’hui clairement retardée et que le mode d’entrée en union s’est également transformé. Les périodes d’union libre sont plus fréquentes parmi les hommes de la jeune génération (1965-1974), certaines célébrations matrimoniales sont aujourd’hui reportées ou annulées et l’ensemble du processus matrimonial a tendance à s’étirer dans le temps. Si, pour la génération la plus ancienne (1945-1954), le premier mariage n’était pas une affaire d’argent, pour les hommes de la plus jeune génération, en revanche, l’obtention d’un emploi rémunéré est cruciale pour la formation d’une première union. Les résultats montrent également que, dans un contexte où les emplois urbains sont de plus en plus informels, on assiste à l’émergence d’inégalités face au mariage entre les jeunes hommes de la jeune génération selon le type d’emploi qu’ils occupent.

PLAN DE L'ARTICLE

• 1 - Worsening job prospects in cities
• 2 - Data and methodology
• 3 - First union occurs later and in a different manner
• 4 -  A more extended process of union formation
• 5 - Effect of employment on first union formation
• Conclusion

Anne-Emmanuèle Calvès « Too Poor to Marry? », Population (English Edition) 2/2007 (Vol. 62), p. 293-311.
URL : www.cairn.info/revue-population-english-2007-2-page-293.htm.
DOI : 10.3917/pope.702.0293.