Several studies have analyzed the optimal design of differentiated VAT tax rates, specific subsidies and empirically evaluated the behavioral effects and welfare implication of specific tax reforms. The results of these studies heavily depend on the behavioral response of consumers, most important the cross-price elasticities of specific goods. In addition, consumption decisions involve a trade-off between today and future consumption. E.g., consuming goods like tobacco, alcohol or fatty foods may provide instant pleasure, but can create negative health consequences in the long run. Likewise, lower investments in education are enjoyable in the youth, but severely affect job, income and wealth opportunities in the future. A large body of evidence suggests that individuals display a high degree of impatience in such situations: they weigh immediate gratification to an extent they regret at a later stage in life. In addition, individuals might suffer from limitations in attention and awareness, e.g., about future consequences of present choices. All these biases may lead people to act against their long-run self- interest and have important implication for the optimal policy design. In particular, these biases may explain why governments largely intervene in private markets, e.g., by making use of additional taxes on unhealthy food (sin taxes), public provision of health insurance (or other forms of in-kind spending), and compulsory schooling. These instruments can differ in their capability to internalize the externalities from today’s consumption on the future self and have different re-distributional consequences.

The task of this sub-project is to extend the previous literature on the behavioral and distributional effects of specific consumption taxes. We focus mainly on three areas, i) the optimal design of consumption taxation; ii) the behavioral responses of consumers to taxes including tax evasion and avoidance; and iii) the distributional and welfare effects of specific tax reforms. In particular, we will theoretically address the question of when price-related instruments (taxes, subsidizes) are Pareto-superior to quantity-based controls (in-kind spending, quantity restrictions) and identify the conditions under which optimal policy mixes should involve more than one class of policy instruments to combat the problems of self- control. Further, we will apply different empirical methods to estimate the behavioral responses of consumers to price and income changes including structural demand models and estimators that exploit regional price variation over time as natural experiments.

To assess the scope for tax avoidance, we will use theoretical, experimental, as well as quasi-experimental approaches and randomized field-experiments. Specifically, we will study to what extent the design of mechanisms that represent new form of tax rebates or nonlinear taxation of consumption might reduce consumers' incentives to collude with sellers to evade taxes. This kind of mechanisms has not been studied sufficiently despite being adopted by policy makers for several decades. Several countries and municipalities, e.g., China, Taiwan, Puerto Rico, São Paulo, have introduced consumer incentive schemes in order to reduce tax evasion by reducing the number of untaxed, i.e., illegal, transactions between firms and consumers. Under these invoice lottery schemes, consumers participate in a lottery if they demand a tax receipt for certain transactions, e.g., restaurant bills. Thus, the chance to win prizes might reduce consumers' incentives to collude with sellers to evade taxes. In previous work, it has been shown that such mechanisms have interesting welfare properties and are able to dampen distortions due to consumption taxation. We will address theoretically and experimentally the capability of such mechanisms to combat tax evasion, taking into account consumer behavior.

Finally, we will draw on previous work focussing on income taxation to study the distributional effects and welfare implications of changes in consumption taxation. For the welfare analysis we will apply a framework for welfare comparison with heterogeneous price effects and heterogeneous preferences of consumers. Moreover, we will investigate the cost (and tax) pass-through incentives of multi-product firms in presence of asymmetric cross-demand effects that might arise from some consumers being uninformed about the prices of some goods or consumers ignoring the prices of some goods when they make their base good purchasing decisions.