« Back to Results

House Price Dynamics and Indexes

Paper Session

Saturday, Jan. 4, 2020 10:15 AM - 12:15 PM (PDT)

Manchester Grand Hyatt, Regatta C
Hosted By: American Real Estate and Urban Economics Association
  • Chair: Chris Redfearn, University of Southern California

Shifting House Price Gradients: Evidence Using Both Rental and Asset Prices

Han Liu
George Washington University


Alonso (1964), Mills (1967) and Muth (1969) formalize the monocentric urban spatial structure model and provide the gradient approach to analyze the issue of sub-urbanization. Urban sprawl has been observed and studied since then. However, recent evidence suggests a reversal: center cities are gaining population. This literature has been based on changes in population density or the slope of the housing asset price gradient. This paper adds to this research by estimating the changing slope of both the rental and asset price gradients for a wide metropolitan areas in the U.S. over a 30 year period. The steepening of the asset and rental gradients is then analyzed at dis-aggregate geographic levels, showing that regions have not all followed the national trend. For the owner-occupied market, asset price gradient has been steepening in all regions but the Midwest. Whereas in the rental market, all regions but the Northeast experienced comparable increases in the slope of the rent gradient. The house price gradient, whether measured using asset or rental prices, increased in slope fastest in big cities with high GDP and large population. Possible reasons for the steepening of house price gradient are then developed using the standard urban model (SUM). The Muth-Mills equation predicts that the slope of house price gradient should vary directly with congestion in urban travel and inversely with housing unit size. The empirical tests evidenced in this paper are performed using a specially constructed panel data set of U.S. metropolitan areas and confirmed these expected relations.

Local Constant-Quality Housing Market Liquidity Indices

Dorinth Van Dijk


The average time on market (TOM) of sold properties is frequently used by practitioners and policymakers as a market liquidity indicator. This figure might be misleading as the average TOM only considers properties that have been sold. Furthermore, traded properties are heterogeneous. Since these features differ over the cycle, the average TOM could provide wrong signals about market liquidity. These problems are more severe in markets where properties trade infrequently. In this paper, a methodology is provided that allows for the construction of constant-quality housing market liquidity indices in thin markets that can be estimated up to the end of the sample. The latter is particularly important since market watchers are generally interested in the most recent information regarding market liquidity and less in historical information. Using individual transactions data on three different types of Dutch municipalities (small, medium, and large) it is shown that the average TOM overestimates market liquidity in bad times and underestimates market liquidity in good times. The option to withdraw is the most important reason why the average TOM is misleading. Furthermore, constant-quality liquidity leads the average TOM and price changes. The indices not only show that illiquidity is higher during busts, but also that liquidity risk is higher. Additional results suggest that setting a high list price relative to the estimated value results in a higher TOM, but this effect differs over time. Both the list price premium and the effect on sale probability are higher during busts. Differences in housing quality over the cycle, however, also play a significant role. Finally, the method allows for the construction of indices that are more robust to revisions, especially in thinner markets.

How Auctions Amplify House-Price Fluctuations

Alina Arefeva
University of Wisconsin-Madison


I develop a dynamic search model of the housing market in which prices, determined
by auction, exhibit greater volatility than prices in the search and matching model with
Nash bargaining from the literature. This helps solve the puzzle of excess volatility of house
prices. The outcomes of the two models dier in hot markets when buyers' house values
are heterogenous. With Nash bargaining, a buyer who gets a house is chosen randomly
among interested buyers, so prices are determined by the average house values. In auctions,
competition among buyers drives up prices to the willingness to pay of the buyer with the
highest value. In hot markets, the highest values uctuate more than the average values,
making the auction prices more volatile than the negotiated prices. This high volatility is
constrained ecient in the sense that the equilibrium allocation decentralizes the solution
of the social planner problem constrained by the search frictions.

The Most Wonderful Time of the Year? Thin Markets, House Price Seasonality, and the December Discount

Erling Roed Larsen
Oslo Metropolitan University
Andre Anundsen
Oslo Metropolitan University


In Norway, house prices tend to drop in December. This regularity could be caused by a composition effect, a seller effect, or a thin market effect. This article exploits a high-resolution transaction data set with exact sell dates to demonstrate the existence of a December discount. To show existence, we deal with unobserved unit and seller heterogeneity using unit fixed effects, ask prices, and appraisal values. We examine generating mechanisms and find that cross-sectional evidence supports a thin market effect. We find no evidence of stressed sellers. Examination of bidding behavior in a bid-by-bid micro auction data set indicates that sellers grow impatient as the holiday season nears. Scrutiny of seller behavior in advertisement data shows reduced activity on the supply side in December.
Shawn Rohlin
Kent State University
Steven Bourassa
Florida Atlantic University
Katherine Kiel
College of the Holy Cross
Ronan Lyons
Trinity College Dublin
JEL Classifications
  • R3 - Real Estate Markets, Spatial Production Analysis, and Firm Location
  • C4 - Econometric and Statistical Methods: Special Topics