The Probability of Sale for Residential Real Estate

Article excerpt


This work investigates the determinants of the probability of sale during a given marketing span for residential properties. The inconsistent empirical relationship between property price and property selling time in numerous prior studies suggests an investigation into the determinants of a successful marketing effort is warranted. Results indicate that marketing time, seller motivation, certain property attributes, and location significantly affect the probability of successfully selling a property. These findings have implications for future works investigating, or relying on, the success or failure of residential property marketing efforts. Additionally, this work provides practical guidance to brokers and sellers of residential property.

The joint determination of residential property price and selling time is well accepted in the real estate brokerage literature. In fact, virtually all recent works that investigate the impact of an identified attribute on property price and marketing time attempt to empirically model the relationship of price and time-on-market simultaneously.1 Anglin, Rutherford, and Springer (2003) use search theory to argue that the relationship between price and marketing time is positive; indeed, this seems to be a fairly widely-held view. However, the majority of the formally reported models from recent works reveal sales price and time-on-market relationships that do not conform to this accepted belief. Additionally, according to Anglin (2006), stability in market conditions must be assumed if the hazard function used to model time-on-market is to provide valid estimates of property marketing time. The lack of a consistent sales price and time-on-market relationship in the reported models and concerns about stability of market conditions combine to warrant an explicit investigation into the determinants of the "failure" being modeled by hazard functions in real estate brokerage research (i.e., the probability of sale).

This study provides an original investigation into the determinants of the probability of a sale. Findings suggest that seller motivation, property attributes for which buyers' and sellers' may have different marginal valuations, and property location significantly affect the outcome of marketing efforts. These findings not only have implications for future research investigating the success or failure of residential property marketing efforts, but also for studies that simply rely on hazard modeling of time-on-market. Additionally, the specific results detailed below provide practical assistance to both real estate brokers and owners wishing to sell their property.

The remainder of the work is arranged as follows: first, there is a review of the existing literature followed by a discussion of the motivation for the study. Second, the data are described then the model and empirical analysis are presented. Finally, there is a summary, along with remarks about future research into determinants of the probability of sale.

Literature and Motivation

Cubbin (1974) and Miller (1978) make what appear to be the first connections between property price and property marketing time. The existence of a relationship between these two metrics is now widely acknowledged. In fact, a number of the hedonic pricing studies reviewed in Sirmans, Macpherson, and Zietz (2005) model price and time-onmarket in the presence of each other. However, in the interest of compactness, the discussion of prior literature here concentrates on more recent works, referred to collectively as the modern pricing and time-on-market literature. By avoiding older studies, the most recently reported relationships between property price and marketing time, as well as the most current modeling technology, can be brought into sharper focus.

The eighteen studies that comprise the modern pricing and time-on-market literature include Sirmans, Turnbull, and Dombrow (1995), Yavas and Yang (1995), Forgey, Rutherford, and Springer (1996), Genesove and Mayer (1997), Glower, Haurin, and Hendershott (1998), Munneke and Yavas (2001), Huang and Palmquist (2001), Rutherford, Springer, and Yavas (2001), Knight (2002), Anglin, Rutherford, and Springer (2003), Allen and Dare (2004), Rutherford, Springer, and Yavas (2005), Allen, Faircloth, and Rutherford (2005), Turnbull, Dombrow, and Sirmans (2006), Turnbull and Dombrow (2006), Turnbull and Dombrow (2007), Huang and Rutherford (2007), and Rutherford, Springer, and Yavas (2007). …