Many social and economic demographers have concluded that socio-economic change in itself cannot account for the classic fertility transition in Europe (nineteenth and early twentieth centuries). The transitions in different regions occurred too quickly once they started and they occurred under too wide a range of social and economic conditions. Furthermore, fertility behavior appears to have been relatively homogeneous across regions with close linguistic and cultural—but not necessarily geographic—proximity (Coale and Watkins 1986 ; Lesthaeghe and Surkyn 1988 ; Watkins 1990). A growing literature on contemporary fertility transitions in developing countries similarly shows that the time path of fertility change is typically different from what can be accounted for by conventional measures of socioeconomic or program change (Bongaarts and Watkins 1996). Moreover, analyses of women's social networks and contraceptive behavior provide a cumulating micro-evidence that the local social environment exerts systematic and important influences on contraceptive and fertility behavior in developing countries that can facilitate or impede the diffusion of modern family planning and lower fertility (see Chapter 3).
These findings have led to an increasing emphasis on social interactions in theories of fertility decline (see Chapter 1). However, specific theoretical analyses of social interactions and fertility dynamics are required to further support this emphasis on social interactions as a potentially important mechanism in facilitating fertility change in contemporary developing countries and in historical Europe. In this and the subsequent chapters we therefore develop a new theoretical analysis that shifts our emphasis from individual fertility behavior to aggregate fertility dynamics in the presence of social interaction. In particular, we consider the feedback loop between the fertility behavior of individuals and that