Marital separation, especially when dependent children are involved, is a significant and disruptive life event that often involves considerable changes in living standards and in how families organise their economic life. International research shows that many people experience a large decline in economic standards of living following separation and that these effects often last many years. Separation is also associated with a significant increase in the risk of poverty. One of the most common findings from the empirical literature is a substantial gender gap in the impact of separation. Most studies find that women, on average, are worse off, especially if they have children. Results for men are more varied with some studies finding a rise in average living standards, and others a decline. At the same time, behind these averages, lies considerable heterogeneity: many men experience a decline in living standards and some women gain from separation. Very little research attention has been directed towards understanding the economic consequences of marital separation in New Zealand. Unlike most European and English-speaking OECD countries there are no New Zealand estimates of the impacts of divorce and separation on incomes or living standards, the gender gap in outcomes, or the extent to which welfare provisions, child maintenance policies and other state measures ameliorate its negative effects. This thesis seeks to extend current knowledge and understanding of economic outcomes of separation for New Zealand families with children. It aims to provide answers to three questions: i. What are the short- to medium-term economic impacts of marital separation for the New Zealand men and women with dependent children included in this dataset? ii. How do economic outcomes compare for ex-partners? iii. What are the likely impacts of the new liability assessment formula in the Child Support Amendment Act 2013 on child support payments and receipts? The thesis includes three related studies. The first uses propensity score matching and difference-in-differences estimation to estimate the short- and medium-term impacts of separation on men’s and women’s total family income and equivalised income. The data for this study comes from a large-scale longitudinal administrative dataset of tax and welfare records. The second takes advantage of the dyadic information in that dataset to examine outcomes for ex-partners relative to each other. The third study focuses on child support policy and, in particular, the likely distributional effects of the recently introduced changes to the liability formula in the Child Support Act.