Abstract We test for gender differences in the real estate appraisal industry. We ask whether female appraisers evaluate different types of properties and whether they value the properties differently compared to their male counterparts. We largely do not find any statistically or economically significant differences between male and female appraisers using a nationwide panel of mortgages for the period 2000–2007. We also find that appraisers' gender does not play a role in appraisal outcomes.
We study the impact of severe health shocks, caused by first-time strokes, heart attacks, and cancer diagnoses, on housing tenure choices in the United States. Using data from the Panel Study of Income Dynamics (PSID), we estimate a value-added model that controls for preshock housing tenure, health status, and an extensive set of individual and household characteristics. We find that experiencing a severe health shock significantly reduces the probability of homeownership by about 2.1 percentage points. This effect remains robust when further controlling for preshock disability, healthcare expenditures, and household fixed effects. Furthermore, we show that our results are mainly driven by preshock owners exiting homeownership after the shock, while there is no effect among preshock renters. When focusing on exits from homeownership, we also find that the effects appear more salient among less affluent, older, and single homeowners. Exploring suggestive evidence for potential mechanisms, we find that health shocks tighten financial constraints by increasing healthcare expenditures and reducing labor supply and labor income.
Many variables involve the modeling of spatial effects, and their dynamics over time. This article presents a linear model in which spatiotemporal random effects are modeled by graph-Laplacians. A graph-Laplacian flexibly encodes adjacency in both space and time, in our case not depending on unknown parameters. The graph-Laplacian can be input for a prior in a Bayesian estimation setup, or used as regularization term in a Ridge regression. A spectral decomposition of the graph-Laplacian significantly reduces computation time for estimation. As an application, we estimate graph-Laplacian hedonic pricing and repeat-sales models on sales prices of Australian residential properties in the period from 1990 to 2024. Bayesian and Ridge regression estimation results are very similar, although the computation time for the Ridge regression is orders of magnitude faster, however at the expense of missing posterior density functions. Our results highlight the advantages of graph-Laplacians for predicting individual property prices and for producing stable, granular price indexes, also in thin markets.
Pricing power in real estate markets can reduce housing supply and redevelopment relative to the social optimum. We show how pricing power interacts with popular redevelopment subsidies and zoning regulations. Using building-level rental income data from NYC, we find that increased concentration is correlated with increased rents. Finally, we use the model to estimate the first building-level housing elasticity, finding that markups account for between one-sixth and one-third of rents in the city.
This paper examines how collaboration between listing and selling agents influences housing market outcomes and how these effects differ across the institutional environments of conventional and foreclosure (REO) transactions. Using more than 150,000 singlefamily sales from the Dallas-Fort Worth metro area, we develop a conceptual framework describing how REO market structure can generate heterogeneous collaboration effects. We estimate jointly determined price and time-on-market (TOM) equations with detailed agent controls and distinguish between repeated interactions within specific agent pairs (depth) and broader patterns of collaborative activity (breadth). In conventional sales, neither dyad depth nor collaboration breadth materially affects prices or TOM once agent characteristics are accounted for. In the REO segment, repeated interactions beyond the first two joint sales are associated with price discounts of roughly two percent, while collaboration has little effect on TOM. Overall, the evidence indicates no meaningful collaboration effects in conventional transactions and modest, price-specific effects in the institutionally constrained REO market.
This article estimates foreclosure discounts in Cape Town, South Africa, where, until 2019, foreclosure auctions occurred without reserve prices. Using newly constructed data linking sheriff auction notices to the universe of property transactions, rich property characteristics, and municipal service-request records, I document large foreclosure discounts that vary by empirical approach. In a cost-adjusted hedonic framework, properties sold at foreclosure auction transact at a discount of 14.5% relative to comparable non-foreclosed sales; repeat-sales estimates yield substantially larger discounts of 30.2%. These estimates bracket the plausible range of the true discount: the hedonic provides a conservative lower bound after netting out buyer-incurred transaction costs, while the repeat-sales provides an upper bound that additionally captures time-varying deterioration not fully observed in cross-sectional data. The magnitudes documented here are large relative to settings with reserve prices, reflecting an institutional environment where the absence of a reserve price leaves auction outcomes effectively unbounded on the downside. The implications are especially severe in South Africa's recourse mortgage system: foreclosed borrowers who lose their homes at auction may also remain liable for any balance between the sale price and the outstanding mortgage, compounding the household-level costs of foreclosure. The findings contribute to an understanding of how foreclosure-price formation operates in a no-reserve-price auction environment, and of the implications of that institutional design for realized prices and homeowner outcomes.
This article examines how spatial variation in housing ages affects income segregation in US cities. We develop a dynamic general equilibrium hybrid Tiebout model with durable housing that combines local public finance and urban land use theories. Heterogeneous households sort across school districts based on their demand for local public goods and within districts based on the trade-off between accessibility and housing rents. Housing is produced by perfectly competitive firms, and each new house lasts for two periods. We find that the age of the housing stock is an important variable that affects the spatial distribution of households, and hence segregation. The cyclic nature of our equilibrium allows us to explain suburbanization and regentrification processes as a result of the endogenously evolving age of the housing stock distribution.
This article leverages the timing and location of Low-Income Housing Tax Credit (LIHTC) properties to study the impact of affordable housing on local public schools. The research design links new developments from 2000-2020 to campus-level demographics, spending levels, teacher counts, and class sizes in a difference-in-differences framework. Through rental unit take-up, LIHTC increases the enrollment of students that income-qualify for school lunch subsidies, with minimal effects on racial composition. Because Federal Title 1 funding is linked to student income, I study heterogeneity in the effect of LIHTC on school spending based on Title 1 status. Schools that opt into Title 1 status when LIHTC arrives experience spending increases, teacher headcount increases, and reductions in class sizes. By contrast, LIHTC schools already designated as Title 1 experience class size increases as per-pupil spending declines. Because most US schools are listed as Title 1, the level of funding for compensatory programming is the key policy lever for schools facing neighborhood change.
We leverage quasi-experimental wildfire smoke shocks to analyze the capitalization of transient air pollution in rent and house prices in Las Vegas from 2008 to 2019. We combine a repeat rent model with an instrumental variable approach, using (i) rental contract rates from newly signed leases, and (ii) the probability of a smoke day based on ground-level wildfire smoke plume history as an instrument for fine particulate matter (). Similarly, we examine a repeat sales model. Our results indicate that increases in daily exposure are associated with decreases in rent and house prices, and these effects are substantially larger when accounting for measurement error. Our findings suggest that tenants value clean air as much as homeowners and make up a key demographic when monetizing the value of clean air to inform environmental policy design.
Do out-of-town (OOT) individual investors suffer from asymmetric information in the rental market? Using single-family residential rental data, we find that OOT landlords charge a 2.85% lower rent than local counterparts, with the discount increasing as the distance between the landlord's address and the property's address increases. Prior local investment experience and stronger social ties mitigate this effect, whereas rental market heterogeneity amplifies it. These findings support that information asymmetry may serve as an underlying mechanism. Our results highlight information friction in real estate rental markets, leading to inefficiencies in rental pricing.
We investigate inconsistencies in property tax assessment regressivity across property classes using public record data from Cook County, Illinois during 2006-2022. Across a battery of tests, we document that assessment regressivity is severe within Class 3 (large residential), Class 5A (commercial), and Class 5B (industrial), whereas assessment ratios are only marginally regressive within Class 2 (small residential). Property Classes 3, 5A, and 5B are significantly more likely to appeal, have successful appeals, and have greater value reductions following a successful appeal, compared to Class 2. The appeal process acts to further exacerbate regressivity within Classes 5A and Class 5B. We calculate the counterfactual tax rate that could have been applied under the hypothetical scenario of zero regressivity within each property class and assuming that assessment ratios equal official assessment rates. The weighted average tax rate across jurisdictions in Cook County during 2006-2022 is 7.8% and the counterfactual tax rate we calculate that produces the same property tax revenue under the above conditions would have been 6.1%.
This study explores the long-term economic effects of a unique form of historical architectural heritage-the city wall-on modern urban spatial structure. Using a Spatial Regression Discontinuity Design, the research compares housing prices on both sides of the Nanjing city wall, one of China's largest and best-preserved urban fortifications. Results show a significant and consistent price premium for properties just inside the walls. Further analysis indicates that this premium is not solely due to the wall's heritage value, nor is it caused by a simple barrier effect or agglomeration dynamics on either side of the boundary. Instead, it stems from deep-rooted path dependence within the city: early investments in the internal road network and the build-up of local amenities, which together maintained a persistent urban structure and produced lasting capitalization effects within the wall.
We test whether new condominium construction generates vacancies in a local housing market through induced moves. Using detailed address-history microdata, we track households who moved into a newly built 512-unit condominium tower in Honolulu, Hawai'i, which included both market-rate and income-restricted units. We identify prior addresses and follow vacancy chains across multiple rounds of moves. The vacated homes were substantially cheaper than the new units and spanned diverse locations and housing types. Income-restricted units produced fewer secondary vacancies, but those vacancies were concentrated at lower price points. Our results show that new condominium construction eases supply constraints and expands local affordability. The distinct filtering dynamics between market-rate and income-restricted units have important implications for inclusionary zoning policies.
Traditional capital structure theories face severe limitations when applied to securitized real estate and real estate investment trust (REIT) markets. Of note, the regulatory environment faced by these firms dramatically alters their economic incentives and limits their ability to self-finance growth and expansion activities. As such, firms in this industry with continuing needs for external capital are uniquely positioned to benefit from reduced valuation uncertainty engendered by enhanced information flow. Against this backdrop, the current investigation examines whether, and to what extent, options market trading intensity serves as a value-relevant, noise-reducing information signal that may be used to inform REIT borrowing and capital structure decisions. Specifically, we document that increased REIT options market trading activity is strongly associated with reductions in overall firm leverage levels through the channel of enhanced equity issuance. Additionally, increased options market trading intensity is associated with a relative increase in the use of unsecured debt and a corresponding reduction in the use of collateralized bank debt and term loans. Importantly, these results appear to be most pronounced within firms that are financially constrained and/or informationally opaque. Taken together, and consistent with predictions derived from pecking order theory, these findings suggest that the enhanced information flow and resulting price discovery attributable to options market activity allow REITs to retain financial flexibility and ensure continuing access to credit.
We estimate the price impact of very nearby concurrently listed properties in the Sydney housing market and assess their competition effects. We apply a hedonic model with spatiotemporal effects regularized via a graph Laplacian prior at the month-by-SA2 regional level to seven SA4 subregions of metropolitan Sydney. The model structure enables localized identification of the effect of nearby active listings, while controlling for local trends in a data sparse environment. We find that an additional active listing within 250 m reduces the sale price by 0.5%-1.3%, of which 0.3%-0.9% reflects the estimated competition effect when compared to the impact of listings further away. In contrast, nearby listings that have been settled show no consistent effect, highlighting that the price discount arises from buyer perceptions of competing alternatives during the active sale window, rather than supply conditions. The effect is strongest in middle-tier regions such as Parramatta and Blacktown, and weaker in the inner core and outer fringe. These findings are relevant to sellers and agents, as very nearby competing listings that are active at the time of sale arise in 10%-25% of cases in Sydney.
Rent control policies have gained renewed legislative momentum in the United States, but are rent-regulated landlords adhering to these policies? Answering this question is critical to understanding the policy's impact. Using a unique panel data set from the New York City Housing and Vacancy Survey (NYCHVS), we investigate noncompliance with rent caps in New York City. We uncover evidence indicative of widespread rent overcharging. During our sample period, over 30% of rent-stabilized apartments without turnover had rent increases exceeding the city's rent caps. Moreover, we find that racial and ethnic minorities are more likely to be overcharged than their White counterparts. Supplemented with building permit and code violation data, we provide evidence that our findings are unlikely to be driven by policy provisions that allow additional rent increases, notably (1) preferential rent and (2) major capital improvement.
In response to the COVID-19 pandemic, numerous countries implemented lockdowns. In Victoria, Australia, a unique two-tier system was employed, segregating areas with a Ring of Steel boundary and imposing additional restrictions within. This study focuses on the impact of lockdowns on housing prices and rents, exploring whether people are willing to pay a premium to live in areas with fewer lockdown restrictions and thus proposing this premium as an alternative measure of lockdown cost. We utilized a spatial difference-in-differences design to test on the lockdown boundary area and address many confounding factors. The research reveals a 7%-8% relative drop in housing rents within the Ring of Steel, dissipating within 6 months after lockdowns ended. A surprisingly large drop of 6%-7% in housing values is observed inside the Ring of Steel. These empirical estimates suggest homebuyers' behavioral biases could further depress housing values during a pandemic.
This study utilizes 1.4 million residential transactions incorporating flood hazard estimates derived from a nationwide elevation model and hydrological data to examine the role of flood protections to mitigate the negative impact of flood risk on home prices. We find that low-rise homes in unprotected flood-prone areas face a 3.5% price discount, whereas mid-/high-rise properties are unaffected. Postinstallation, homes near flood protections experience price increases of 10.7%, with no significant difference across home types. This premium mirrors the preprotection discount observed for the same homes. Robustness checks using manual flood risk assessments confirm our findings, offering valuable insights for various stakeholders.
Previous research demonstrates that housing prices frequently move in tandem across regions, underscoring the interconnectedness and correlation present within housing markets. Building on this foundation, our study advances the analysis by examining quantile co-movements and the synchronization between local and national housing markets. Using the quantile factor model across the full distribution of housing prices, we identify distinct factor structures at the lower and upper tails that contrast with those observed in the middle of the distribution. This analytic framework enables the detection of previously hidden factors influencing housing markets. With this approach, we illuminate how housing price dynamics interact across market segments, price levels, and geographic areas. Our findings reveal that co-movements can vary substantially across low, stable, and high housing price regimes, thus providing more comprehensive and nuanced economic insights into the complex nature of housing price fluctuations.
This article examines how the leasing activities, contract features, and pricing of the Class A office leasing market have evolved since 2019 across five major US markets: Los Angeles, the Bay Area, Dallas, Washington, DC, and New York City. Using a granular dataset of 73,508 office leases from 2010 to 2024, we find a broad-based contraction in leasing volume and meaningful adjustments in contract features, including increased reliance on free rent and shifts in tenant-improvement usage. More importantly, we document structural changes in the determination of net effective rents at the lease level. In several major markets, longer leases, which were previously associated with rent discounts, began to command premiums after 2019, indicating a revaluation of contractual duration. We also find intensified spatial polarization and substantial reordering of tenant industry rent premiums, suggesting increased segmentation across geography and industry.
