Fertility Decline in Bangladesh: Evidence from the 1993-94, 1996-97 and 1999-2000 Bangladesh Demographic and Health Surveys (BDHS)

Barkat-e- Khuda, ICDDR,B: Centre for Health and Population Research
Roy Nikhil, ICDDR,B: Centre for Health and Population Research
Nirod Ch. Saha, ICDDR,B: Centre for Health and Population Research

BACKGROUND: Bangladesh has experienced considerable decline in fertility, with the total fertility rate (TFR) dropping by almost half from 6.3 in 1971-75 to 3.3 in 1997-99. The sharp decline in fertility has, indeed, attracted the attention of population experts around the world. While some have attributed the fertility decline in Bangladesh to a strong family planning programme, others have argued that fertility decline in Bangladesh cannot be solely attributed to the role of the family planning programme but also to other major socioeconomic changes. DATA AND METHODS: This paper is based on national level data for the period 1975-2000. For the multivariate analysis of the determinants of fertility decline, data from the 1993-94, 1996-97 and 1999-2000 Bangladesh Demographic and Health Surveys (BDHS) have been used. The BDHS employed a nationally representative, two-stage cluster sample design. In the 1993-94, 1996-97 and 1999-2000 BDHS, a total of 9,640, 9,127, and 10,544 eligible women (ever-married women aged 10-49 years) respectively were interviewed. For the multivariate analysis of the determinants of fertility decline, we have considered only currently married women of reproductive age: 8,842, 8,306 and 9,540 respectively from the 1993-94, 1996-97, and 1999-2000 BDHS. The dependent variable, fertility, is a dichotomous variable, and refers to either no children born or one or more children born during the reference period (three years preceding the survey). The independent variables included desire for more children, mass media, ever use of family planning, education, employment status, possession of land, and electricity connection. Logistic regression has been used for the analysis. RESULTS: During the 1970s and 1980s, Bangladesh had high fertility, and most people believed that the high fertility trend would continue for several years into the future. The total fertility rate (TFR) was 6.3 in 1971-75, and stayed at a high level of 5.1 even after a decade in 1984-88. Between 1984-88 and 1989-91, the drop in fertility became noticeable, with the TFR dropping to 4.3 in 1989-91, or about a child less, on average, per woman than about five years ago. Since then, fertility decline has been quite rapid, with women, on average, having one fewer child in 1991-93 compared to that in 1989-91. However, the pace of fertility decline slowed thereafter, and plateaued at 3.3 during the 1994-2000 period. Women, on average, have higher fertility in rural areas (3.5) than urban areas (2.5). Among the six geographic divisions of the country, fertility is lowest in Khulna Division (2.7), while it is highest in Sylhet Division (4.1). Education has a strong effect on fertility. Women with at least some secondary schooling have, on average, 2.4 children, which is 1.7 fewer children than those with no schooling (4.1). Using data from the 1993-94, 1996-97 and 1999-2000 Bangladesh Demographic and Health Surveys, logistic regressions were carried out to examine the effects of selected independent variables on fertility, the dependent variable (measured as the probability of not having given birth for the three years preceding the surveys). The results show that the probability of not having given birth during the reference period was considerably lower among women who said that they would like to have a child after two years than those who said that they wanted to have a child within two years. Also, it was lower among women who said that they wanted no more children than those who said that they wanted to have a child within two years. Those with access to mass media had higher probability of not having given birth than those without access to mass media. Those who had ever used family planning methods had higher probability of not having given birth than those who never practiced contraception. Women with education, especially those with secondary schooling, had higher probability of not having given birth than those with no schooling. The probability of not having given birth was higher among working women than among housewives. Women belonging to landed households had higher probability of not having given birth than those belonging to landless households. The probability of not having given birth was higher among women belonging to households with electricity connection, an indicator of not only economic condition of the household but also its level of modernization, than women belonging to households without electricity connection. CONCLUSIONS: Bangladesh presents an excellent example of a country where substantial fertility decline has taken place, despite the fact that it is one of the poorest country in the world in terms of per capita income. Certainly, the success of the family planning programme has been largely instrumental in bringing about the fertility decline in Bangladesh. The programme efforts themselves have largely been facilitated by considerable socio-economic changes taking place in the country over the past decade and a half. Such changes include education, access to mass media, urbanization, rising landlessness as well as declines in the desired family size and infant and child mortality. Therefore, any satisfactory explanation of fertility decline in Bangladesh would have to take into account the role of both the family planning programme and the socioeconomic changes that have been underway in Bangladesh.