The mixed effects of income upon successive births may explain the convergence phenomenon.

Summary This paper finds that an increment of income has a more positive (less negative) effect on the probability of a family having another child when it consists of relatively few children than of relatively many children. That is, within a group of people whose income is increased, the increase may be expected to influence the average family with no children or one child to increase family size, and to influence the average family that would otherwise have many children to have fewer. To put it another way, increased income reduces the variance of family size across families. This provides an explanation for the 'convergence' of American families to two to four children. The basis for these conclusions is a cross-sectional analysis of white women in the U.S. Census of 1960.