نقش هنجارهای اجتماعی در یک مدل از ازدواج و طلاق

Abstract In the US, the rate of divorce has increased at an alarming rate since the 1960s. This paper presents a mechanism which gives rise to the emergence of multiple equilibria and a discrete jump in the rate of divorce in a simple search environment. We attempt to show that social norms influence the way each agent searches for a matching partner and hence the probability of divorce. In a low-divorce equilibrium, agents are willing to spend more cost in the search process. As a result, the act of divorce becomes a more accurate signal of unobservable characteristics of a divorced agent. This type of social norms is self-consistent in equilibrium as social stigma attached to divorce is relatively high and this forces agents to be more discreet in the search process. For exactly the opposite reason, a high-divorce equilibrium can also be supported where agents are willing to spend little cost in the search process. In the light of this logic, the rapid increase in the rate of divorce can be seen as a movement from a low-divorce to a high-divorce equilibrium, possibly triggered by a temporary shock.

1. Introduction There is a world-wide trend that the rate of divorce has increased drastically over the last several decades. In the US, for instance, the rate of divorce had approximately doubled between 1963 and 1974.1 This increase turned out not to be just a temporary shock: the rate of divorce has been consistently high since then. Moreover, the rapid increase in the rate of divorce is not an isolated incident in the US. Many western countries have seen a sharp increase in the rate of divorce as well. Since the stability of marriage has profound welfare implications, much attention has naturally been paid to the causes of this change. Indeed, it does not appear overly difficult to make a list of potential causes as virtually all western countries had gone through several structural changes during that span. For instance, it is often argued that the inception of no-fault divorce laws in many western countries contributed to this increase as, under no-fault divorce, the cost of obtaining a divorce decreases. Of course, the fact that no-fault divorce was introduced when the rate of divorce started to increase is by no means indicative of the direction of causation. It is certainly plausible that the easing of the divorce process could be the effect of social changes as well as the cause of them. Actually, Michael (1988) points out that the increase in the rate of divorce preceded the changes in the divorce law in many US states. Allen (1998) argues that the introduction of no-fault divorce was motivated by a rise in the number of inefficient marriages throughout the 1950s and 1960s.2 The magnitude of the effect on the rate of divorce is still somewhat unclear.3 Another possibility is that changes in the economic situation triggered this increase as the stability of marriage is apparently closely associated with many economic factors.4 As Becker et al. (1977) argue, an increase in women’s wages tend to decrease the value of marriage as it gives women more viable outside options. Then, it is certainly arguable that the recent increase in women’s wages and labor force participation might contribute to this increase. For illustrative purposes, we use those economic factors as potential originating causes although it should be emphasized that the paper is not primarily concerned with the causes of this change themselves. While there is little doubt that unexpected changes in the economic situation make otherwise acceptable options unacceptable, it is still not so clear how this sudden and rather substantial change was triggered by the gradual and continuous increase in women’s wages and labor force participation over time. To account for this change, then, what is missing is a mechanism which links those potential causes to a catastrophic rise in the rate of divorce. Given this fact, the main goal of the paper is to show the existence of multiple equilibria in a simple matching framework with search friction. By multiple equilibria, we mean a situation where two different rates of divorce arise simultaneously as an equilibrium outcome.5 To this end, the paper focuses on social norms in the search process as the main force behind this structural change, although we certainly do not intend to imply that other factors play little role in this. We argue that the presence of different social norms in the search process could give rise to the emergence of multiple equilibria, one with a high rate of divorce and the other with a low rate of divorce. In the presence of multiple equilibria, the rapid increase in the rate of divorce can be seen as a movement from a low-divorce to a high-divorce equilibrium. The existence of multiple equilibria has a serious policy implication as simply eliminating those potential causes may not affect the aggregate rate of divorce. The gist of the paper is intuitively simple. In general, the average quality of divorced agents is not as good as that of the entire population. Precisely because of this fact, agents are punished when they get divorced. The degree of punishment, however, depends on public perception about the quality of divorced agents which in turn depends critically on social norms in the search process.6 That is, the cost of divorce is endogenously determined in our model. As social norms force agents to take more cost in the search process, public perception about the quality of divorced agents tends to become worse as the act of divorce becomes a more accurate signal of undesirable characteristics. This type of social norms is self-consistent in equilibrium: agents are indeed forced to spend more cost in the search process as the cost of divorce is high. When agents do not spend much cost in the search process, on the other hand, the act of divorce becomes more of a fuzzy signal. This is also self-consistent in equilibrium: agents do not need to spend much cost in the search process as they are not punished as harshly when they get divorced. We show that these two different social norms can coexist under some plausible conditions. The paper proceeds as follows. In Section 2, we outline the basic environment of the model. In Section 3, the equilibrium conditions are analyzed and the possibility of the existence of multiple equilibria is discussed. Finally, Section 4 concludes.

4. Conclusion Marriage is a process of satisficing. As it is nearly impossible to find a perfect match, agents are virtually forced to set the minimum standard which is acceptable. We argue, however, that setting the minimum standard is a collective decision problem. The driving force of the model is that the amount of effort spent in the search process is negatively correlated with the expected gain when the match is not successful. The expected gain in the remarriage market, for instance, decreases as agents become willing to take more cost in the search process. The presented model depicts a situation where two types of equilibrium coexist in a simple search environment. The stability of matches goes through a catastrophic change when an economy moves from one equilibrium to the other, possibly triggered by a temporary shock. This implies the the probability of a match being successful in general, and the rate of divorce in particular, may exhibit hysteresis where a temporary shock has a persistent effect by changing the rules of the search process. The existence of multiple equilibria implies that simply eliminating originating causes may have no effect on the rate of divorce. As a final note, the presented model has clear relevance to the labor market.15 The cost of job turnover is high when potential employees are willing to spend more cost in the search process: changing jobs is often regarded as reflective of one’s own disabilities since job mismatches are rare in this situation. This perception can be supported in equilibrium as it indeed forces potential employees to spend much cost in the search process. The cost of job turnover is low, on the other hand, when potential employees are willing to spend little cost in the search process: changing jobs is often regarded as resulting from job mismatches in this situation. This perception can also be supported in equilibrium as potential employees do not need to spend much cost in the search process. Different social norms in the job search process may help explain cross-country differences in the job turnover rate and the natural rate of unemployment.