Regional fertility differentials in Brazil: a decomposition method application

Abstract

Total fertility rate (TFR) is the most widely used fertility measure and provides up-to-date information on levels and trends. However, changes in the mean age of childbearing can lead to mistaken if not considered in the analyzes, especially in low fertility contexts. Changes in the mean age of childbearing between two periods affect – increasing or decreasing – the observed fertility, even in the absence of other changes. Defined as tempo effect, the determination requires the decomposition of the TFT variation. This paper aims to measure the tempo effect in Brazil, regions and states, in 2000 and 2010, from the application of the Bongaarts and Feeney model (1998). The model measures a pure level (quantum), free from variations in the mean age of childbearing, from which it is possible to determine the tempo effect. The 1991, 2000 and 2010 Demographic Censuses are used. The results indicate that, in 2000, in general, low fertility levels were inflated by the anticipation of births; in 2010, fertility levels across the country were further reduced by the tempo effect.