[Pre-transitional fertility in Latin America: a forgotten subject]. / La fecundidad pre-transicional en America Latina: un capitulo olvidado.

PIP: Although Latin American fertility over the past three decades has been fairly well documented, less is known of pretransition fertility, operationally defined as occurring before 1960. Data shortcomings have limited fertility estimates for these years, especially for population subgroups. This document attempts to reconstruct levels and trends in the total fertility rate for the pretransitional period using census retroprojection. The study was done in rural and urban zones to evaluate the existence of heterogeneity in each country, and in accordance with the availability of census data. Four countries were selected to represent the diversity of observed fertility in the region. Honduras's transition was late, slow, and initiated at a high level of fertility. Even in 1990 the total fertility rate exceeded 5. Costa Rica and Colombia had intense transitions and 1992 total fertility rates of around 3. Colombia had a rapid transition and total fertility rates averaging under 3 in 1990. The general hypothesis of this work is that pretransition fertility varied between and within countries. It was not precisely "natural" fertility inasmuch as the elites practiced some fertility control. Fertility change began when groups practicing fertility control became a majority, starting with the development of the middle class. The estimates indicate that pretransitional fertility in the 4 countries was not higher than the level at the moment when the transition began. Pretransitional fertility was relatively stable through about 1950, with variations in the total fertility rate not exceeding 1 child per woman. The total fertility rates of the 4 countries already differed before the 1950s, calling into question the view of the population of Latin America as uniform and homogeneous during the period. Heterogeneity was not limited to the lower fertility of Argentina and Uruguay on the one hand contrasting with high rates elsewhere in Latin America. Substantial differentials were observed in all 4 countries in rural and urban rates. Only in Honduras were the rural-urban fertility differentials caused almost entirely by differing patterns of nuptiality; in the other 3 countries differentials persisted even after controlling for earlier and more stable rural unions. Mortality declines and increases in fertility apparently occurred in all 4 countries in the 1950s and early 1960s. The secular reduction in fertility began after the increases in all countries studied except Honduras. In Colombia and Costa Rica, the fertility increases during the 1950s were primarily due to changes in nuptiality, but in Chile less than one-fourth of the fertility increase was due to nuptiality change. Part of the fertility increase was also due to improved health conditions.^ieng