Um estudo sobre a mortalidade dos aposentados idosos do Regime Geral de Previdência Social do Brasil no período de 1998 a 2002

Abstract

The ageing process of the Brazilian population, initially on account of the fertility reduction and more recently reinforced by an increase in longevity, has been generating important implications for the public policies such as the pension policy. This occurs because even in the absence of any modification on the retirement age, there is an increase on the average time lived by people in the in the retired status. Despite of the various efforts put on the challenge of identifying the Brazilian mortality pattern, studies that aim to analyze specifically the mortality curve of the Brazilian elderly, which would be highly useful for retirement calculations, are scarce.The general objective of this dissertation is to study the elderly population retired in the Brazilian General Regime of Social Pension between 1998 and 2002. As an specific objective, it has been built a mortality table based on probabilities of death of these retired people observed in the relevant period of time.Our study is based on the administrative registre data of the Social Pension Processing Data Enterprise. Therefore, our database contains dependable information, since for one to concede or cease benefits they will need to assess a series of documents what causes to reduce or avoid considerably some mistakes.The mortality curve for the retired elderly in the Brazilian General Regime of Social Pension from 1998 to 2002 was analyzed by sex and age and the life expectancy originated from the curve was compared to the ones obtained from the tables of IBGE, 2007 and AT-83. This comparison is important to show the applicability of the mortalilty curve presented in this dissertation.Life expectancies originated from the probabilities of death calculated with data from the Brazilian General Regime of Social Pension are very close to the ones obtained on the IBGE-2007 and AT-83 tables. From the age 72 up until the 92 one curve practically overlap the other.The greatest prerogative of the table calculated in this dissertation lies in the fact that since it was estimated based on the administrative date, the numerators and denominators come from the same source and the data is collected directly from official fonts, therefore avoiding mistakes such as coverage mistakes and digit preference. These pieces of information are collected for financial purposes, frequently under a strict control of Government agencies, thus far less prone to inaccuracy.