| Models of welfare participation that include both the welfare participation decisions of individuals and the selection decisions of welfare administrators are developed. An individual's welfare decision is based on budget constraint variables, personal and household characteristics, province of residence, and the unemployment rate. A welfare administrator's selection decision is based on an applicant's characteristics and past income, province of residence, and province-related variables. Pooled cross-section time-series (monthly) data from the 1988--90 Labour Market Activity Surveys (LMAS) are employed. The results show that the estimated coefficients on the wage rate, welfare tax rate, and welfare benefit are generally consistent with the theoretical predictions of the standard labour supply model. Estimates of the coefficients associated with several variables indicate an important role for administrative selection.;The role that welfare administrators play in welfare and work participation and in the selection of participants for employment and training (ET) programs is studied using survey data for Alberta. These ET programs are expected to increase participants' earnings and their likelihood of working and staying off welfare. The results show that age, education level and family type are significant determinants of ET program participation. However, ET program participation does not have a positive impact on participants' earnings, although it has a positive impact on participants' likelihood of working. Only long term ET program participation is found to have a positive impact on participants' likelihood of staying off welfare. |
