Using a Canadian-American Natural Experiment to Study Relative Efficiencies of Social Welfare Payment Systems

We study whether social welfare recipients may end up paying more for their grocery if social welfare payments are more concentrated over time. We first present a theoretical model showing that lower incomes in general and a lower lower bound of the income distribution lead to less mobility for poorer consumers. This causes local stores to have more market power and increase their prices when the incomes of poorer people go down and/or when the number of poorer people goes up. Secondly, we verify these theoretical findings by using a natural experiment to study links between food prices and the more restrictive timing of social welfare payments in Montreal, Canada compared to the timing in Bangor, Maine. We find some statistically significant evidence of : i) a negative effect on prices in the week of social welfare check issue ; ii) increasing prices over a month. We also find that some socio-economic factors such as a higher percentage of single-parent families in one area may increase prices charged by grocery stores in that area.