Do financial incentives affect rural physician supply? Empirical evidence from a Canadian experiment

Rural physician shortages remain a persistent challenge in health workforce planning, and financial incentives are commonly used to improve geographic physician distribution. This study exploits Saskatchewan's Rural Physician Incentive Program (RPIP), introduced in 2013, as a natural experiment to evaluate the impact of direct annual payments for rural practice on the rural-urban distribution of family physicians. Using administrative physician counts from the Canadian Institute for Health Information and population data from Statistics Canada, we construct health region-area type-year panels for short-run (2009-2017) and long run (2009-2021) windows, comprising 72 and 104 observations, respectively. We first document distributional patterns using location quotients, and then implement a two-way fixed-effects difference-in-differences (DiD) design comparing rural and urban areas within mixed health regions (regions containing both rural and urban populations) of Saskatchewan before and after policy introduction. To strengthen identification, we conduct multiple robustness checks, including a triple DiD framework that incorporates Manitoba as a comparison province, selected for its geographic proximity, similar rural-urban population structure, and absence of a comparable contemporaneous incentive program. Distributional results indicate that rural physician rates remained persistently below urban rates throughout the study period. Main DiD estimates show no statistically significant narrowing of the rural-urban physician rate gap within mixed health regions of Saskatchewan following the program implementation. The Triple DiD estimates similarly fail to detect a significant policy effect relative to Manitoba. Taken together, the results provide no evidence that financial incentives altered geographic physician maldistribution. The findings suggest that, depending on size and scope, stand-alone incentives may be insufficient to meaningfully address family physician maldistribution without integrated, multi-component workforce strategies.