The broad, long-term objective is to exploit a uniquely detailed data set to investigate the causes of fertility decline in Prussia from 1875 to 1910. The analysis will examine in greater depth the issues raised by the Princeton European Fertility Project. The Prussian data cover a population in excess of 40 million (12.5 percent of all of Europe) and refer to geographical units some one-fifteenth the size of the European Fertility Project's. The proposed analysis of Prussia includes not only much more detailed socioeconomic data, but also data which are geographically and temporally comparable. Therefore these data will support a more appropriate test of fertility transition theory in general and will allow a more appropriate test of the main conclusions reached by the European Fertility Project in particular. Those conclusions are that no socioeconomic theory can account for European fertility decline and that particular. Those conclusions are that no socioeconomic theory can account for European fertility decline and that cultural and regional influences appear paramount. Our preliminary analysis of a subset of Prussian data already suggests that those conclusions are incorrect, and that analysis at a finer level of geographic detail reveals important within-region variation. This general aim will be pursued by carrying out the following step: 1. Collect and enter into machine-readable form a major new data set covering social, economic, political, religious, and demographic data from census and vital registration sources for about 550 Prussian Kreise (districts), about 100 large cities, and thirteen districts in the city of Berlin from 1861 to 1914. 2. Develop from these data a set of demographic measures of total fertility, marital fertility, non-marital fertility, nuptiality, infant mortality, net migration, and gross in- migration. 3. Develop from these data set of geographically and temporally comparable socioeconomic measures for each Kreis and each large city. These include measures of religion, linguistic group, income, certain prices, occupational composition, availability of banking and insurance services, extent of communications and transportation infrastructure, quantity and quality of children's schooling, urbanization, city size, farm size, and voting behavior. 4. Use pooled cross-section time series methods with these data to investigate the causes of fertility decline and its timing in Prussia from 1875 to 1910, allowing fixed or persistent effects for each area. 5. Model and estimate pooled cross-section time series methods incorporating interactions between diffusion effects and socioeconomic development effects. We believe that this will be a more comprehensive test of fertility transition theory within the European context than has previously been possible. Much of this theory is derived from studies of fertility decline in contemporary less developed countries. As a consequence, many of our variables are identical to those used in such studies. Some of our variables are more appropriate theoretically and cover a relatively longer time span. Lack of data for important variables combined with a lack of comparable data over time has often forced researchers to rely on simple cross-sectional analyses of fertility decline in Europe and in LDC's leading in turn to possibly erroneous conclusions. We expect our research to shed light on the mechanisms involved in the determinants of both historical and contemporary fertility level, onset of fertility decline, and pace of fertility decline.
| Lee, R D; Galloway, P R; Hammel, E A (1994) Fertility decline in Prussia: estimating influences on supply, demand, and degree of control. Demography 31:347-73 |
