طلاق یکجانبه در مقابل حضانت فرزند و حمایت از کودکان در ایالات متحده

Abstract This paper explores the response of the divorce rate to law reforms introducing unilateral divorce after controlling for law reforms concerning the aftermath of divorce, which are omitted from most previous studies. We introduce two main policy changes that have swept the US since the late 1970s: the approval of the joint custody regime and the Child Support Enforcement program. Because those reforms affect divorce decisions by counteracting the reallocation of property rights generated by the unilateral divorce procedure and by increasing the expected financial costs of divorce, it is arguable that their omissions might obscure the impact of unilateral divorce reforms on divorce rates. After allowing for changes in laws concerning the aftermath of divorce, we find that the positive impact of unilateral divorce reforms on divorce rates does not vanish over time, suggesting that the Coase theorem may not apply to changes in divorce laws. Supplemental analysis, developed to examine the frequency of permanent shocks in US divorce rates, indicates that the positive permanent changes in divorce rates can be associated with the implementation of unilateral divorce reforms and that the negative permanent changes can be related to the law reforms concerning living arrangements in the aftermath of divorce. This seems to confirm the important role of these policies in the evolution of divorce rates.

. Introduction In an article in the American Economic Review, Wolfers (2006) finds that reforms in the divorce law in the United States (US) during the 1960s and 1970s had a positive effect on divorce rates. Wolfers claims that this result does not back up the applicability of the Coase theorem to marital relations since divorce rates are not neutral to changes in divorce laws. 1 However, he also observed that the effect was transitory; after a decade, no effect on divorce rate could be discerned. 2 This generates doubts about the empirical evidence that does not support the predictions of efficient Coasian bargaining. To explain his puzzling results, Wolfers suggests that a situation where spousal bargaining was close enough to the efficient one – consistent with the Coasian approach – can account for the small and transitory effect estimated. In this paper, we provide an alternative explanation by presenting evidence that later reforms that introduced changes in divorce settlements may explain the diminished effect of unilateral divorce on the divorce rate. Two primary aspects of law are relevant to divorce and both may affect divorce decisions (Fine and Fine, 1994). First, there are laws that regulate how spouses obtain a divorce, and these include the unilateral divorce regime.3 Second, there are laws that govern the living arrangements in the subsequent periods after divorce, including such matters as spousal support, child support, and child custody.4 These are not included in Wolfers (2006) but they may have significance in the evolution of the divorce rate.5 Although, from a theoretical point of view, it can be suggested that those changes in divorce settlements have an ambiguous effect on divorce (see Nixon, 1997, Rasul, 2006a and Halla, 2011), previous empirical research has found that both changes in the financial obligations of parents and the introduction of joint custody negatively affect divorce rates (Nixon, 1997 and Brinig and Buckley, 1998).6 Thus, it is arguable that the analysis of one of those aspects of law relevant to divorce alone might in some way obscure the impact of unilateral reforms on divorce rates. This is even more relevant in the US since while the share of the population covered by the no-fault unilateral reforms increased from the late 1960s, reaching 50% of the population in the early 1970s (see Fig. 1), a trend of reforms occurred in the area of post-divorce child custody and child support. Empirically, it is unclear whether the dummy variables included by Wolfers (2006) to capture the dynamic response of divorce only pick up the path of the adjustment of divorce rates to unilateral divorce. Wolfers observes that the effect of unilateral divorce law reforms on divorce rates had dissipated a decade after the implementation of the unilateral divorce law, which coincides with the rise in the incidence of joint custody (Fig. 1). The timing of both reforms differs by at least a decade in almost all states in which those reforms were implemented (Friedberg, 1998 and Leo, 2008). In the area of child support, the US Congress approved several laws to try to ensure child support payments. The main reforms were the Child Support Amendments of 1984, the Family Support Act of 1988, the Child Support Recovery Act of 1992, the Personal Responsibility and Work Opportunity Reconciliation Act of 1996, and the Deadbeat Parents Punishment Act in 1998 (see Sorensen and Halpern (1999) for a review of state statutes). This again tallied with the time at which a negative response of divorce rates to divorce law reforms was found in Wolfers (2006). We argue that what lies beneath Wolfers's results are two countervailing forces, which together produce the observed pattern. Thus, the initial increase and subsequent decrease in divorce rates would be the response of divorce rates to the initial changes in divorce laws followed by custody reforms and Child Support Enforcement (CSE) efforts. Coverage and timing of reforms. Fig. 1. Coverage and timing of reforms. Figure options Initially, we include the reforms that govern the aftermath of divorce into Wolfers's specification using data on the divorce rate from 1956 to 1988. We introduce both child custody law reforms and CSE efforts. Our results suggest that the long-run effect of divorce law reforms on the divorce rate observed by Wolfers may be the result of both unilateral reforms and changes in the aftermath of divorce. When we separate both effects, we find evidence of a persistent impact of divorce laws on divorce rates, although these results are sensitive to the inclusion of state-specific trends. This is robust to a range of alternative specifications and to the selection into marriage effect. These findings suggest that the Coase theorem cannot be applied to marital dissolution. As an additional check that the changes in the aftermath of divorce are driving our findings, we separate the analysis into divorcing couples with and without minors in order to check whether the behavior of childless couples – the sub-population not affected by legal changes in the aftermath of divorce when they obtain a divorce – is driving our results instead of the reactions of couples with minors. We present additional evidence suggesting that the joint custody law and the reinforcement of child support predominantly affect the exit from marriages of couples with minors as opposed to changing the divorce pattern of childless couples. Finally, since even after adding the reforms on the custody laws and child support to the analysis it is unclear whether divorce laws have a persistent effect on divorce, we explore the frequency of persistent shocks in US divorce rates by exploiting another technique, a time-series analysis.7 The advantage of this methodology is that it lets data “speak for themselves”, (see Piehl et al., 2003 and Kuo, 2011); this allows us to test whether and when there have been changes in divorce rates without imposing any a priori timing (such as the dates of the reforms). We analyze three possible scenarios (for a review of the literature on structural breaks, see Perron, 2006). First, the divorce rate is stationary. In this scenario, the divorce rate is basically stable; after a shock, such as divorce law reforms, short-run effects on the divorce rate would be observed, but in the long run, the divorce rate should return to its equilibrium level. In the second scenario, divorce is stationary around a process that is subject to structural breaks. In this setting, occasional shocks cause permanent changes in the equilibrium rate itself, but most shocks only cause temporary movements of the divorce rate around the equilibrium level. The third scenario consists of the divorce rate exhibiting a unit root. In this case, all shocks have permanent effects on the level of divorce.8 The clear result of the time-series analysis is that not all shocks have transitory effects on the divorce rate. This result is robust to a number of alternative tests. There is no single scenario to identify the behavior of the divorce rate; we find empirical evidence of stationarity around a process that is subject to structural breaks, where only a few occasional shocks have permanent effects, and of unit root, with all shocks having a permanent effect on the divorce rate. In addition, our results suggest that persistent positive changes can be associated with major changes in divorce laws and those permanent negative changes can be related to changes in custody laws and the CSE program, since the break dates and the dates of the reforms are close to each other. The present paper is organized as follows. Section 2 discusses the results of Wolfers (2006). In Sections 3 and 4, we introduce custody law reforms and CSE efforts into Wolfers's analysis. Section 5 includes the supplemental analysis of the frequency of permanent shocks in divorce and gives possible explanations for these changes, and Section 6 concludes.

Conclusions This paper aimed to disentangle the effects of law reforms that govern the aftermath of divorce from the effects of unilateral divorce in determining the behavior of US divorce rates. Because empirically it is unclear whether the coefficients measuring the response of divorce rates to divorce law reforms are only capturing the adjustment path of divorce rates to unilateral divorce when it is omitted major reforms that have swept the US since the late 1970s, we introduce to the analysis of the impact of unilateral divorce two main reforms in the area of post-divorce: the adoption of the joint custody regime and the CSE program. The incorporation of the custody law change is important since the possibility of joint custody may counteract the reassignment of property rights generated by the unilateral divorce reforms, according to the Coase theorem. Under joint custody, parents have to collaborate and cooperate in decisions affecting the child; this implies a return to a situation in which mutual consent is necessary. It is not possible to leave out of this analysis the CSE program either. The increase in the efforts to try to ensure child support payments is relevant to the study of the response of divorce rates to divorce law reforms when less restrictive divorce laws are correlated with stricter enforcement efforts made by the states in order to achieve the objective of reducing child poverty and welfare costs. Our results suggest that the divorce rate increased immediately after the adoption of unilateral divorce as in Wolfers (2006). After a decade, two countervailing forces seem to be operating. We show empirical evidence indicating that the negative evolution of the divorce rate since the 1980s seems to be because of law reforms concerning the aftermath of divorce rather than the reverse response of divorce rates to the implementation of unilateral divorce laws. Further, some of our estimates point to a permanent impact of the unilateral divorce reforms on divorce rates, suggesting that the Coase theorem cannot be applied to marital relations. All in all, we view our results as evidence in favor of the important role of laws that regulate the aftermath of divorce, but we also believe that a more thorough examination of the mechanisms through which those reforms operate is an interesting question for future research. We also developed a supplemental analysis to explore the frequency of permanent shocks in US divorce rates. A clear finding from this analysis is that not all shocks have transitory effects on divorce rates, which is robust to a range of alternative tests. This result can be interpreted in the context of evaluating the effects of divorce laws on divorce rates. The positive permanent changes in divorce can be associated with the implementation of unilateral divorce and the negative permanent changes can be related to the reforms in the laws that regulate the aftermath of divorce, again suggesting an important impact of divorce law reforms on the evolution of divorce rates.