Working Paper: NBER ID: w20084
Authors: Wojciech Kopczuk; David J. Munroe
Abstract: Houses and apartments sold in New York and New Jersey at prices above $1 million are subject to the so-called 1% "mansion tax" imposed on the full value of the transaction. This policy generates a discontinuity (a "notch") in the overall tax liability. We rely on this and other discontinuities to analyze implications of transfer taxes in the real estate market. Using administrative records of property sales, we find robust evidence of substantial bunching and show that the incidence of this tax for transactions local to the discontinuity falls on sellers, may exceed the value of the tax, and is not explained by tax evasion (although supply-side quality adjustments may play a role). Above the notch, the volume of missing transactions exceeds those bunching below the notch. Interpreting our results in the context of an equilibrium bargaining model, we conclude that the market unravels in the neighborhood of the notch: its presence provides strong incentive for buyers and sellers in the proximity of the threshold not to transact. This effect, the identification and recognition of which is novel to this paper, is above and beyond the standard extensive margin response. When present, unraveling affects interpretation and estimation of bunching estimates. Finally, we show that the presence of the tax affects how the market operates away from the threshold---taxation increases price reductions during the search process and in the bargaining stage and weakens the relationship between listing and sale prices. We interpret these results as demonstrating that taxation affects the ultimate allocation in this search market.
Keywords: Mansion Tax; Transfer Taxes; Real Estate Market; Bunching; Market Efficiency
JEL Codes: H21; H71; R31
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| mansion tax (H24) | bunching of transactions just below the $1 million threshold (F38) |
| mansion tax (H24) | distortion in price distribution (D39) |
| mansion tax (H24) | unraveling effect (D85) |
| bunching of transactions just below the $1 million threshold (F38) | increased price dispersion (D49) |
| mansion tax (H24) | inefficiencies in the search process (D83) |
| mansion tax (H24) | gap in the distribution of sales prices (D39) |