Abstract
In the last decades, female permanent sterilisation became the most used method of contraception in Mexico. During this time, the demand for pills, condoms and other short-term contraceptives fell consistently. The shift in the demand for contraceptives raises concerns among demographers that the timing of children may remain unchanged regardless of observed reductions in period fertility rates. This paper assesses such ideas in the context of the timing of a first child using duration models as the main analysis tool. Findings suggest that young cohorts of women are effectively delaying first birth relative to the experience of older generations.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
The total fertility rate (TFR) is a measure of the number of children that a woman would have at the end of their fertile life if she follows the current ‘typical’ fertility behaviour at all stages of her life (it can be expressed in terms of children per 1,000 women). Age-specific fertility rates indicate the number of birth per 1,000 women in different age-specific groups.
Cases where the given dates of birth of mothers and their children implied a negative duration interval were excluded. The analysis is done conditional on this selection.
During estimation, it was found that moving the calendar limits of the generation dummies for +5 and −5 years did not result in important changes in the estimated parameters and their standard errors.
The underlying continuous time hazard θ(.) is the model the researcher would estimate if continuous time duration data were available. However, if one collects duration intervals that only change in a discrete manner (say, years) and x i is constant between t and t+1, then θ(.) can be written as h it . In the duration data literature, h it is known as the discrete time hazard function (see, for instance, Meyer 1990; Sueyoshi 1995).
Notice that the Extreme Value distribution of the discrete time hazard h it is a direct consequence of the proportional hazard functional form of the underlying continuous time hazard θ(t). To avoid confusion between h it and θ(t), h it is referred to as the ‘EV hazard’ whenever θ(t) is supposed to belong with the proportional hazard family.
The Extreme Value type I distribution may not be attractive in applied work because it has a fat right tail (i.e. the distribution is skewed to the right and has skewness to 1.13955). Clearly, for large samples, central limit theorem arguments would suggest that a rather symmetric distribution such as Normal is appropriate.
To simplify exposition, from now on, the reader should understand ‘discrete time’ hazard anytime the hazard function is referred to.
The typical individual was found to be Catholic and had 5.35 years of education at age 12.
According to the ENADID, between 1992 and 1997, single mothers contributed 5% of the most recent pregnancies that resulted in live births. No single mothers are reported among non-Catholic women who were born between 1953 and 1957. In contrast, the generation of non-Catholics who were born between 1968 and 1972 contributed 8.41% of children born to single mothers. In other words, non-Catholic women in the young cohort contributed a higher proportion of out of wedlock births than non-Catholic women in the older cohort. To complete the picture, data from the ENADID show that while 78% of women in the 1953–1957 cohort married before age 25 (or entered a consensual union), only 69% of women in the 1968–1972 cohort married before that age. Clearly, these descriptive statistics support the hypothesis that, at least in relative terms, the Catholic Church is succeeding in persuading young couples to delay marriage and sexual intercourse.
References
Arulampalam W, Stewart M (1995) The determinants of individual unemployment in an era of high unemployment. Econ J 105:321–332
Bloom D, Trussell J (1984) What are the determinants of delayed childbearing and permanent childlessness in the United States? Demography 21(4):591–611
Bumpass L, Rindfuss R, Janosik R (1978) Age and marital status at first birth and the pace of subsequent fertility. Demography 12:75–86
Cabrera G (1994) Demographic dynamics and development: the role of population policy in Mexico. Popul Dev Rev 20:105–120
Chen R, Morgan S (1991) Recent trends in the timing of first births in the United States. Demography 28(4):513–533
CONAPO (2001) La poblacion de Mexico en el nuevo siglo. Document downloadable at http://www.conapo.gob.mx
CONAPO (2002) National population programme 2001–2006
Elbers C, Ridder G (1982) True and spurious duration dependence: the identifiability of the proportional hazard model. Rev Econ Stud 49(3):403–409
Gomez J (1996) La fecundidad y el crecimiento de la descendencia. Demos 9:8–10
Gustafsson S (2001) Optimal age at motherhood. Theoretical and empirical considerations on postponement of maternity in Europe. J Popul Econ 14(2):225–247
Happel S, Hill J, Low S (1984) An economic analysis of the timing of childbirth. Popul Stud 38(2):299–311
Heckman J (1979) Sample selection bias as an specification error. Econometrica 47(1):153–162
Heckman J, Singer B (1984) The identifiability of the proportional hazard model. Rev Econ Stud 51(2):231–241
Heckman J, Hotz V, Walker J (1985) New evidence on the timing and spacing of births. Am Econ Rev 75(2):179–184
INEGI (1999) National survey of demographic dynamics 1997
INEGI (2000) Estadisticas historicas de Mexico
INEGI (2001a) Estadisticas demograficas cuaderno no. 13
INEGI (2001b) Sistema de indicadores para el seguimiento de la situacion de la mujer en Mexico. Information available at http://www.inegi.gob.mx
Lindstrom D (1998) The role of contraceptive supply and demand in Mexican fertility decline: evidence from a microdemographic study. Popul Stud 52:255–274
Meyer B (1990) Unemployment insurance and unemployment spells. Econometrica 58(4):757–782
Mier y Teran M, Rabell C (1990) Introduccion: la transicion demografica en la decada de los ochenta. Rev Mex Sociol 52(1):3–13
Narendranathan W, Stewart M (1993) How does the benefit effect vary as unemployment spells lengthen? J Appl Econ 8:361–381
Newman J, McCulloch C (1984) A hazard rate approach to the timing of births. Econometrica 52(4):939–961
Schmidt P, Witte A (1989) Predicting criminal recidivism using split population survival time models. J Econom 40:141–159
Sueyoshi G (1995) A class of binary response models for grouped duration data. J Appl Econ 10:411–431
Welti C (1997) Cambios en la fecundidad. Demos 10:16–18
World Bank (2001) World development indicators
Zavala de Cosio E (1989) Dos momentos en la transicion demografica. Demos 2:6–7
Zavala de Cosio E (1990) Politicas de poblacion en Mexico. Rev Mex Sociol 52(1):15–32
I am grateful to Wiji Arulampalam, Mark Stewart, John Ermisch, Jeremy Smith and two anonymous referees for useful comments. I am also grateful to the National Council for Science and Technology (CONACYT) for its financial support.
Author information
Authors and Affiliations
Corresponding author
Additional information
Responsible editor: Junsen Zhang
Rights and permissions
About this article
Cite this article
Miranda, A. Are young cohorts of women delaying first birth in Mexico?. J Popul Econ 19, 55–70 (2006). https://doi.org/10.1007/s00148-005-0046-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00148-005-0046-7