Employment and the Nonlinear Relationship of Household Income on Divorce in Iran Using the Quasi-Panel Data Logit Model

In this paper, the effect of income distribution on divorce in Iran has been investigated using micro data of Household Expenditure and Income Survey of urban households in 2014 and applying the Quasi-panel data method. Primary data included 18885 urban households. At first, sub-section of the data were selected including divorced and married male and female (16503 Households) and clustered in provinces. The data was then analyzed based on the initial and categorized data. In the next step, using the Deaton (1985) Quasi-panel data, Logit model is estimated with the Maximum likelihood method. The results of first estimation indicate that household per capita income reduces the probability of divorce, and the result of the secondary model (the initial model with including the square of per capita income) indicates a U shape effect of per capita income on the probability of divorce. That is, the probability of divorce in the two groups of income (low and high) is higher than the middle income holders. Threshold per capita income is estimated equivalent to 1275 thousand Toman on average and monthly for urban households or 5100000T in a month for a family of four in 2014. Also, having job reduces the probability of divorce. Therefore, unemployment reduction policies and supportive policies to reduce income inequality in society will be a way to reduce the probability of divorce.

Employment and the Nonlinear Relationship of Household Income on Divorce in Iran Using the Quasi-Panel Data Logit Model Keywords:

Divorce , Household Per Capita Income , Logit Method , Quasi-Panel Data , Micro Data of Household Expenditure and Income Survey