By Divya Singh
In recent decades, significant increases in the rent-to-income ratios have worsened the affordability of living in cities across the world. A widely promulgated solution is to build more housing in cities. But if there is an excessive demand for housing in cities, why don’t developers build more units themselves? Experts have identified strict land-use regulations as one plausible reason for why they do not. Another reason is the property tax, which, by taxing any improvements on the land, reduces the return on new investment, which in turn restricts housing supply. A plausible policy solution would be to provide a tax break on new residential investment. But this raises the question: who benefits from such a tax break?
In my job market paper, I estimate who benefits locally from the property tax exemption for new housing in New York City. I find that the property tax exemption increased the number of rental housing available. However, at the local level, a larger number of tax-exempt housing within a block distance from existing buildings increased their rents by 2%. At first, this is a surprising result because it suggests that this increase in local housing increased rents locally. However, I show in my paper that this is not surprising when we consider the additional effect of the new investment, or the amenity effect. In particular, new available housing attracted high-income tenants whose residence in the area increased attractive business amenities such as sidewalk cafes and restaurants. The increased number of businesses had a positive spillover effect on rents in nearby buildings. In other words, the arrival of new residential building spurred gentrification.
To provide empirical evidence in support of my conclusions, I leveraged a fascinating policy experiment in New York City, which increased the rental housing supply by more than 1 percent within a year. The policy change reduced the property tax benefits associated with 421a tax exemption in select regions of the city (blue areas in Figure 1). The 421a tax exemption exempts new residential investment from property taxes for 15 or 25 years[RM1] . The implementation of this policy change was delayed for more than a year. For some developers, this meant they could choose to either move their investment ahead in time or else pay higher property taxes in the future. One consequence of this delay was a massive spike in rental housing starts just before the deadline (See Figure 2). I calculated there were an additional 1% rental housing starts as a result of the anticipated property tax increase of about 2.5%.
To calculate the rent effects in the neighborhoods which received new residential investment, I collected and harmonized several datasets from city government agencies. The final result was a dataset of investments, rents, and taxes at the level of a building in New York City, which is an improvement over previous studies that relied on data at a broader geographic level. Next, I compared the rent growth in the existing buildings located near vacant parcels (and which are more likely to receive new investment nearby) to the rent increases in existing buildings located in the same census tract but not near a vacant parcel at the time of the policy reform (and hence less likely to receive new investment nearby). I found that a 1% increase in the local rental housing stock increased rents in nearby existing buildings by 1.8% (Figure 3). Additional empirical strategies and robustness checks point to a consistent finding that rents increased in existing buildings near the site of new construction.
To make these mechanisms clearer, I then investigated the changes in the demographics and businesses in the affected regions. I found that the number of high-income, college-educated, and white tenants increased in the tracts that received at least one project, compared with those census-tracts which received none. Additionally, there was an increase in the number of businesses (such as restaurants) catering to high-income residents.
The investment onslaught permanently altered some neighborhoods in the city. The case of Brooklyn is particularly impressive. Neighborhoods in Brooklyn selected for treatment, compared with neighborhoods not selected for treatment, saw an increase in sidewalk cafes (which, we have seen, are associated with high-income customers) and a decrease in the laundromats (which associated with low-income tenants) following an increase in the new housing units. Existing building-owners invested more in their buildings and were more likely to apply for permits for renovation in regions which received more investment.
While the results detailed in my paper do not quantify the aggregate effects, they do show that the local effects of additions to a particular locale’s housing stock do not necessarily mirror the aggregate effects. The gentrification effects could partially explain why there is often local opposition to new housing. Existing low-income residents who do not value amenities brought-in by high-income residents are financially worse-off, which I calculate to be about $500 per unit per year. It is worth noting, though, that the welfare loss would be lower if the inflow of high-income residents increased any of those public amenity which are desired by low-income residents. An example of such an amenity is public safety. Whether there are indeed such effects from new residential investment is a topic for future research. Overall, the results highlight that it is crucial to incorporate such spillover effects when designing tax policies that encourage new residential investment.
Singh, Divya (2019). Do Property Tax Incentives for New Construction Spur Gentrification? Evidence from New York City. Link
Divya Singh is a sixth-year Ph.D. candidate in Economics at Columbia University and expects to defend her dissertation in May 2020. Her research spans topics in public Economics, urban, and development economics. Broadly, her research interests include designing tax policies that help tackle issues in growth and inequality. She has worked on topics such as property taxes and consumption taxes. Divya was born and raised in Delhi, India. Before joining Columbia, she received her BA in economics from the University of Delhi and an MA in Quantitative Economics from the Indian Statistical Institute, Delhi.
The Robert Schalkenbach Foundation did not support the production of the research described here, but would like to thank the author for sharing their efforts with us.