The main aim of this thesis is to make a forecast of the future share of single mothers and fathers in Norway, and to identify the factors that influence the probability of men ending up as single fathers compared to the probability of women ending up as single mothers after dissolution of parental unions. In order to make this forecast, I have used the macro simulation program LIPRO (LIfestyle PROjections) version 4.0, which is based on the methodology of multi-state demography. This is a macro model that simulates groups with certain characteristics. In real life, it is the individuals, or the households, that behaves in a certain manner.
Since the household projection contains only a few demographic factors at the macro level, there are not many possibilities for a thorough analysis. The various sensitivity variants computed in this thesis are meant to investigate the consequences of higher or lower parameter values for the processes that determine the proportions of single mothers and single fathers. In other words, my strategy is to analyse how sensitive changes in relevant LIPRO-rates are for the share of single mothers and single fathers in Norway. This gives me the answer to the following two questions; 1. Which of the components of change, or processes, is of the most importance to the share of single mothers and fathers? and 2. How trustworthy are the forecasted changes in the future shares of women/men being single parents?
In chapter 2, I give an overview of the factors that influence parental disruption. Many of them are outside of LIPRO’s reach, as they are aspects of individual behaviour. If one is interested in analysing such behaviour, individual data must be collected. Chapter 3 gives a rather non-technical presentation of the LIPRO model, and in chapter 4, the data used for this thesis is presented.
Chapter 5 includes a household projection for Norway for the years 2002-2032, which indicates a continuation of trends that have been observed for some decades already; fewer couples with children, more cohabiting couples, and more single-person households.
Next, a household projection for the men and women in the age-group 30-49 years being single fathers and mothers for the years 2002-2032, or my Benchmark simulation, is presented. The Benchmark simulation shows that the share of single mothers in the age group 30-49 years decreases, while the share of single fathers in the age group 30-49 years increases. Here, I divide the age group in two, and find that the greatest contributions come from the “younger” women and the “older” men. I also construct a female/male ratio for the shares, and find that from having five times as many mothers as fathers in 2002, this is expected to fall. In 2032, the female/male ratio is expected to be 3.
Further, 8 different sensitivity computations carried out in order to investigate how sensitive changes in LIPRO-rates are for the proportion men or women in the age-group 30-49 years being single fathers or mothers, are presented. The central results here are that pair dissolution seems to be more important than pair formation, that dissolution of consensual unions is more important than the dissolution of marital unions and that marriage is less important than cohabitation. These results agree with other findings in the literature. As the simulations consist of counterfactual changes, and these sensitivity analyses give the same pattern for the female/male ratio as the Benchmark-simulation, I suggest that the change in the ratio and thus also in the shares are real.
The last part of chapter 5 contains two types of sensitivity variants that consist of counterfactual changes, computed in order to investigate the robustness of the household projection. I find that the forecasted changes in the shares of single mothers and fathers are real.