Empirical essays on behavioural responses to taxation

This thesis addresses empirical questions on the behavioural response to taxation. The first essay develops a model of contributions to tax-preferred savings accounts. Contributions to tax-preferred savings accounts are typically constrained by a contribution limit. I develop a simple life-cycle model in which tax filers exhibit "use-it-or-lose-it" behaviour. This connects current contributions to future contribution limits, which leads to the result that an increase in contribution limits can decrease contributions. Using a panel of Canadian tax filers I find empirical evidence that provides support for the model---larger future contribution room is associated with smaller contributions.;The second essay studies further questions on tax-preferred savings accounts. I find that taxes play a role in the decision to contribute to a Registered Retirement Savings Plan, but the effect of taxes is smaller than suggested by the existing literature. I also find evidence that the sensitivity of participation to future marginal tax rates increased after the introduction of the carryforward. This is consistent with the model presented in the first essay.;The final chapter looks at tax incentives and fertility. This essay exploits a unique reform in the province of Quebec that paid benefits following the birth of a new child. I implement two approaches to estimation. First, I use a quasi-experimental comparison of treatment and control groups to estimate the effect of changes in policy. Second, I take advantage of the policy reforms to construct an instrument that enables me to address the endogeneity of fertility to income-tested benefits. I find a strong and robust effect of fiscal incentives on fertility. In an extension, I use this policy variation to form an instrument to study the effect of family size on female labour supply, finding results consistent with the existing literature.