Journal of Housing Economics最新文献

This paper documents stylized and empirical facts associated with co-borrowers in the US mortgage market since the early 1990s. The share of mortgages with a co-borrower has declined dramatically across different income and demographic groups. We show that this decline, despite being a universal phenomenon across the US, evinces significant regional heterogeneity which contributes to the divergence in local mortgage markets outcomes. Regions with a lower co-borrower share have higher mortgage default rates. Further, in an event of an adverse shock, regions with a low share of mortgages with a co-borrower experience persistently lower house price growth, and lower purchase and refinance mortgage growth.

Land prices are at the heart of urban economics but are generally not observed directly. Though they are central to household and firm location choices, land-only sales in urban areas are rare and often outliers. Indeed, urban areas are in part defined by a largely contiguous area of high land-use intensity – those places in which developable land is scarce. In this paper, we make use of more-common market data to infer land prices: house sales. Using locally weighted regressions, we estimate the value of a standardized structure across two urban counties: Maricopa, Arizona and Sedgwick, Kansas. Because the value of the standardized structure should be invariant across different locations in a metropolitan area, any remaining variation in the value surface should reflect differences in land values. By pinning down this surface using vacant lot sales at the periphery, we are able to extract land values throughout the metropolitan area, even in locations where vacant land sales are rare.

Transfer activity in the U.S. market for mortgage servicing rights has increased in recent years. Incumbent servicers are at an informational advantage relative to potential buyers of these servicing rights, introducing the possibility of adverse selection. This paper marks the first investigation of adverse selection in the market for mortgage servicing rights. Using data from mortgage servicers, we find that loans with higher ex ante measures of prepayment and default risk were more likely to experience a servicing transfer. Results from an ex post analysis in which we condition on these risk measures reveals that loans that experienced a servicing transfer were more likely to prepay and default, a finding that suggests that the market for servicing rights is characterized by adverse selection.

Many goods and services have accompanying costs that are not salient at the moment of purchase. Existing research suggests that consumers are inattentive to such costs when making small purchases. There is less evidence about attention to costs associated with large purchases. This paper examines residential real estate transactions and studies the extent to which sale prices adjust to ownership costs. The results are inconclusive, neither ruling out full price adjustment nor lack of price adjustment. Despite the inconclusive result, the inability to decisively rule out incomplete price adjustment to predictable ownership costs (which is suggestive of inattention) is noteworthy, given the high financial stakes of buying a home.

We construct land values for each parcel in Maricopa County (Phoenix), Arizona, from 2000 through 2018 using a novel public dataset containing the universe of land sales and parcel records in the county. We then compare residential land values constructed using two classes of source data, vacant land sales and land under existing structures. Between 2012 and 2018, estimated land values for developed parcels are, on average, 14% higher when estimated using vacant land due to plattage effects and other unobserved factors. Growth rates are similar, facilitating the use of vacant land price indices to trace valuations over time from an accurate base year valuation. Dynamics between prices of Maricopa County land and housing suggest hypothetical land value tax revenues are more pro-cyclical than property tax revenues, with elasticities of 3.3 and 2.3 with respect to national house prices, respectively. By 2018, houses had recovered 96% of pre-crisis (2007) values, but land had only recovered 66%. These findings demonstrate a source of risk of dependence on public revenues from land value taxes versus a base-period revenue-neutral property tax.

We develop a statistical learning model to estimate the value of vacant land for any parcel, regardless of improvements. Rooted in economic theory, the model optimizes how to combine common improved property sales with rare, but more informative, vacant land sales. It estimates how land values change with geography and other features, and determines how much information either vacant or improved sales provide to nearby areas through two levels of spatial correlation. For most neighborhoods, incorporating improved sales often doubles the certainty of land value estimates. Relative to conventional estimators, our method mitigates problems from excess variance and sample selection.

One approach to land valuation, particularly used by accessors, is to base price estimates of target properties on comparable properties that recently sold. These properties are chosen to be close matches of the target unit and their transaction prices are used to predict the market price of the target unit. But the choice of comparables is typically not consistent and transparent. In this study, a systematic analytical procedure for choosing comparables that is easy to implement is developed. A hedonic regression using these comparables is then run and the predicted value of the target unit is the assessed value. One of the advantages of this procedure is that it should be straightforward for assessors and public finance officials to use and understand and easy to explain to residents.

This approach is used to estimate land value as well as market prices (that includes the value of the structure) for single family residential properties using data from Maricopa County Arizona from 2007-2018. The best estimators obtain a median prediction accuracy error of 10% and more than 60% of these predictions have a prediction accuracy error within 10% for later years in the sample. These are within the bounds obtained by Zillow for 666 U.S. counties. Market prices are disaggregated into structure, lot, and neighborhood values. On average, these three components make up approximately 30%, 20%, and 50% of total average price. This provides for a nice “rule of thumb” for decomposing the market average property value into these three components; two of which relate to land value.

Teardowns provide direct information on land values in fully developed urban areas because such properties are valued only for their land and location rather than for the characteristics of the structure. We use two approaches to estimate land values. The first approach is a Stein-like procedure that uses teardown properties and makes efficient use of limited data when a group of variables – in this case, the structural characteristics – is expected beforehand to provide little explanatory power. The second approach is based on an unconditional expectation for the pooled data set of teardown and non-teardown sales. We use data from Chicago and Maricopa County to demonstrate the two approaches. These approaches help to estimate land values in built-up areas where vacant land sales are uncommon and unrepresentative of land values in the market.

Accurate estimates of land values on a property-by-property basis are an important requirement for the effective implementation of land-based property taxes. We compare hedonic, residual, and matching techniques for mass appraisal of residential land values, using data from Maricopa County, Arizona. The first method involves a hedonic valuation model estimated for transactions of vacant lots. The second approach subtracts the depreciated cost of improvements from the value of improved properties to obtain land value as a residual. The third approach matches the sales of vacant lots with subsequent sales of the same properties once they have been developed. For each pair, we use a land price index to inflate the land price to the time of the improved property transaction and then calculate land leverage (the ratio of land to total property value). A hedonic model is estimated and used to predict land leverage for all improved properties. We conclude that the matching approach is the most promising of the methods considered.