Benjamin M. Miller
Graduate Student – Department of Economics
"The Effect of Child Care Tax Subsidies on
Household Expenditure" with Prof. Kevin
Mumford. Krannert Working Paper 1261
This paper examines how consumers respond to a change in a personal income tax provision when there are interactions with other elements of the tax code which makes the financial implications less salient to the taxpayer. We use data from the Consumer Expenditure Survey to provide evidence that taxpayers responded to the 2003 expansion of the Child and Dependent Care Credit without considering important interactions. Taxpayers who only considered the 2003 change to this tax credit would have perceived that the after-tax price of child care had decreased. However, we show that for many low-income taxpayers, the after-tax price of child care actually increased due to interactions with other elements of the tax code, particularly the increase in the value of the Child Tax Credit. Using a difference-in-differences estimation strategy which exploits the heterogeneity in the size of the perceived and actual change in the value of the Child and Dependent Care Credit, we find strong evidence of a child-care expenditure response to the perceived change and no evidence of a response to the actual change. Through falsification exercises we rule out several alternative explanations and interpret the effect as causal. This evidence implies that the low salience of the personal income tax can be used to induce a taxpayer response without providing any actual financial incentives.
Does Validity Fall from the Sky? Observant Farmers
and Exogenous Rainfall
Rainfall shocks, or deviations from mean rainfall, are commonly treated as random variables and have long been the textbook example of an instrumental variable. However, this paper suggests that individuals can form rational and accurate expectations of these "random" deviations, and provides empirical evidence of such behavior in the crop choices of Indian farmers. A variety of implications are discussed. First, this provides a new and unique framework for thinking about the impacts of climate change on crop equilibria. Prior research relied on long-run cross-sectional comparisons to estimate new equilibria, while more recent works use 30-year historical "regime switching" weather variation within regions to estimate transition paths in the medium-run. This approach opens the door for short-run panel structure analysis of the impacts of changes in weather and climate. If individuals are able to predict some of the annual variation in seasonal rainfall, then their response to these expectations can be informative about marginal responses which occur along the path of response to climates change. Second, policy-relevant evidence explains why only certain types of irrigation can reduce the described sensitivity of crop acreage to short-run weather expectations. Finally, this calls into question the validity of deviations from mean rainfall as an instrument for farmer income and other outcomes. This paper presents an extended model of rainfall which identi es biases in the prior use of rainfall shocks and suggests improved methodology for using rainfall as a source of exogenous variation.