Working Paper: NBER ID: w18040
Authors: H. Spencer Banzhaf; Wallace E. Oates
Abstract: We re-evaluate two forms of fiscal illusion in local public finance: debt illusion and renter illusion. The Ricardian Equivalence Theorem for local governments suggests the form of finance of a public program (tax or debt finance) has no effects on substantive outcomes. For the local case, this results from the capitalization of local fiscal differentials into property values. We show that this version of the model is quite restrictive. In particular, in the U.S, context, where state and local interest is exempt from federal taxation, rational behavior may be inconsistent with Ricardian equivalence if local governments can borrow on more favorable terms than individuals. We also suggest a new test for renter illusion (or the renter effect). In particular, whether or not renters are more likely to support public investments in general, the renter effect suggests that renters are more likely to support them when financed with property taxes than with sales taxes. Using data from hundreds of open space referenda in the U.S. using a variety of finance mechanisms, we find evidence that households do prefer debt financing to tax financing, but find no evidence of the renter effect.
Keywords: fiscal illusion; Ricardian equivalence; local public finance; debt financing; renter effect
JEL Codes: H3; H4; H7; Q2; R2; R5
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Household preferences for financing mechanisms (G51) | Support for public investments (H54) |
| Debt financing (G32) | Higher approval rates for referenda (D72) |
| Renter effect (R21) | Support for property tax financing (R51) |
| Communities with higher share of renters (R21) | Support for expenditures for open space (H76) |