Abstract
This paper introduces newly coded information describing province- and year-specific variation in work requirements, diversion, earning exemptions, and time limits. This new information reveals a large decline in the chance of welfare participation of at least 1.1 percentage points (9.2% relative to the unconditional mean rate of participation) associated with stringent combinations of those four new welfare reforms, even after controlling for benefit levels, eligibility requirements, province-specific GDP growth and unemployment. These results replicate previous findings based on aggregate data and extend them with controls for individual-level characteristics. Microdata with individual-level characteristics enable estimates of the effects of new welfare reforms on 46 subpopulations, suggesting that immigrants, native Canadians, single parents and disabled people were far more effected by provinces' aggressive new attempts to limit welfare participation than other Canadians receiving social assistance.