Content area
Full text
LEORA FRIEDBERG * Since early 1996, at least eight states have begun to consider rolling back unilateral and no-fault divorce-a major reversal of the liberalizing trend in divorce laws which began around 1970.' The proponents of tightening the divorce regime often argue that making divorce more difficult will strengthen families. However, both theoretical and empirical evidence is decidedly mixed over whether the "no-fault revolution" actually contributed to the sharp increase in divorce rates in the United States observed over the last 30 years. Thorough empirical work on the subject by H. Elizabeth Peters (1986) was followed by a comment by Douglas W. Allen (1992) and a reply by Peters (1992) . Using the same cross section of women observed in 1979, Peters found no role for the switch to unilateral divorce in explaining rising divorce rates, while Allen found a significant impact. The differences in their results centered on controls for geographical heterogeneity in divorce propensities. In this paper I address the dispute by using a panel of state-level divorce rates. The sample includes virtually every divorce in the United States over the entire duration of the law changes and allows thorough controls for heterogeneity in divorce propensities across states and over time-which turn out to be crucial to the results. Including state-specific trends allows unobserved state divorce propensities to trend linearly and even quadratically over time and reveals that unilateral divorce raised divorce rates significantly and strongly. An additional difficulty in this literature arises in defining precisely what is unilateral divorce. Peters and Allen disagreed whether separation requirements qualify as unilateral divorce. Margaret F. Brinig and F. H. Buckley (1998) questioned whether divorce in states that retain fault grounds for property settlement can be considered unilateral. The empirical analysis in each of those papers retained a binary classification of divorce laws, yet the impact of different types of unilateral divorce can be tested. I find here that the type of unilateral divorce matters. States that adopted divorce laws which were more strongly unilateral had greater increases in the divorce rate. Nevertheless, the evidence shows that adopting any type of unilateral divorce raised the divorce rate.
I combine both sets of results to arrive at final estimates of the impact of unilateral divorce. The estimates...