Factors behind global fertility development after 1950: a multivariate analysis of 128 countries
Lutz, W.
Demographische Informationen 1984: 21-48; 175
1984
PMID: 12339590 Document Number: 405667
Time series of selected socioeconomic indicators of 128 countries representing 97.4% of the world population in 1975 are related to gross reproduction rates (GRR) in various models of bivariate and multivariate analysis assuming linear as well as logistic functional relationships. The data base stems mainly from UN publications and special attention is given to China. For the pooling of time series and cross-sectional data, countries are grouped according to geographical and cultural criteria, and variables accounting for these regional effects are included in the equations. Special emphasis is placed on the development of an analytical framework trying to combine aspects of economic and sociological fertility analysis which account for shortterm economic determination as well as for changes in the system of social norms and in the degree to which these norms are forced on individual behavior. Among other findings, there is a pronounced dichotomy between more and less developed countries with respect to most variables but especially so with fertility. There are those countries with GRRs above 2.0 and below. The explanatory values of the models of multivariate analysis are generally very high with the RZs ranging from 0.73-0.97, depending on the weighting used and on the specification of the model. Mutlicollinearity was reduced by transformation and aggregation of variables. Female life expectancy at birth seems to be the single most important variable in explaining differential fertility; however, no direct causal link may be assumed; instead life expectancy can be seen as a very general indicator of health conditions and quality of life which in turn influence fertility development. Female educational status is the 2nd most important variable. Measures of female educational status relative to men also supports the argument that female social status is relevant for fertility. A positive income effect on fertility appears for the gross domestic product per caput in a multivariate setting which is not significant, however, like the proportion of the population in agriculture, the other economic variable that was included. Religion appears to be significant in explaining differential fertility, even after accounting for all the other variables. Concerning the importance of culture-geographical country groups, it appears that in Europe, North America, and China, fertility has the tendency to be lower than would be expected from the level of socioeconomic development when comparing all the countries--in the Islamic world and in some Central American countries, fertility tends to be higher than expected from socioeconomic standing. In a longitudinal comparison, it seems that from 1950-75, cultural factors became increasingly important in explaining different national fertility levels, whereas the importance of economic development diminished.