Substitutability among Federal Income Tax Deductions: Implications for Optimal Tax Policy

This paper studies households making joint decisions about two tax-deductible activities: charitable giving and paying interest on home equity lines of credit (HELOC). By allowing for interactions, either substitutability or complementarity, between deductions, I provide a fuller understanding of the elasticity of taxable income (ETI) and the tax-price elasticity of charitable giving. The tax-price of paying HELOC interest depends on the tax schedule as well as current interest rates. Therefore, the two activities have different tax-prices and the identification of their interaction is possible. Using the Survey of Consumer Finances, I find that the level of each activity falls with its own tax-price and rises with the other activity’s tax-price. I show that in theory the ETI is a weighted sum of such own and cross tax-price effects between activities. I then apply this theory and my estimates to illustrate the trade-offs of three base-broadening tax policies. A conventional policy which removes a subset of deductions while keeping other deductions intact cannot guarantee either more tax revenue or greater efficiency. In contrast, an optimal policy would lower all deductions by differing degrees. However, designing an optimal policy requires knowing the values of all own and cross tax-price elasticities between deductions. Instead, I recommend a second-best policy of “uniform partial deductibility” which lowers all deductions to the same degree; it can guarantee improvement without prior knowledge about elasticities between deductions. I am grateful to Zachary Bethune, Raj Chetty, R. Morris Coats, John Friedman, Don Fullerton, Wayne-Roy Gayle, Damon Jones, Sarah Reber, Patrik Sandas, Dan Silverman, Steven Stern, Sarah Turner, the participants of the University of Virginia applied microeconomics research group, Bankard Applied Micro Workshop, UVA-Richmond Fed Research Jamboree and Southern Economic Conference, and especially Leora Friedberg and John Pepper for very helpful comments and discussions. I thankfully acknowledge financial support from the Bankard Fund for Political Economy. All errors are my own.