Determinants of Time on the Market in a Thin Real Estate Market

Abstract

In the paper we investigate the factors that affect a property's time on the market (TOM) in the residential real estate market in the case of a thin, illiquid market, such as those found in Central and Eastern European (CEE) countries. In contrast to liquid markets, the time it takes to sell a property can vary from a few days to a few months. In Slovenia a residential property's marketability depends strongly on the price dynamics of the market, housing characteristics and the degree of overpricing. The most important determinants of the marketing time are the cost and availability of housing finance as well as the housing price index, and all three can be directly related to the affordability of housing. We develop a two-stage model of the determinants of TOM and a duration model. The results of the list price model show that as the age of property increases, its (list) price decreases with diminishing pace, while the size of the property has the opposite dynamic. The presence and size of parking lots also has a positive effect on the (list) price. The results of the Time on the Market model show that property characteristics, market conditions and macroeconomic determinants are all statistically significant determinants of TOM. The degree of overpricing turned out to be a statistically significant determinant of time on the market. However, this effect does not seem to be statistically significantly non-linear (U-shaped). Higher house prices (at the national level) and the average interest rate on housing loans both extend a property's time on the market while better availability of housing loans, in contrast, shortens the TOM. We additionally estimated a proportional hazard model of the TOM that yielded consistent results.

Classification and Selection of Tenants in Residential Real Estate: A Constructivist Approach

“…This is of relevance, because the real estate market not only has an undeniable importance for economic development, but is also a cornerstone of other economic activities: the real estate market provides shelter to other domestic services (Goodhart & Hofmann, 2007). Although signs of recovery from the crisis are beginning to emerge, the unique characteristics of the conjuncture it created in the real estate market cannot be overlooked, namely: a lack of liquidity; heterogeneity of assets; and decreasing sales in almost all segments of the real estate industry (Cheng, Lin, & Liu, 2008;Cirman, Pahor, & Verbic, 2015). In the residential rental market, in particular, understanding these unique characteristics is important to ensure minimum rent/price volatility, regardless of the economic conjuncture.…”

Cluster-based econometric analysis of house prices in Slovenia

“…Th is eff ect becomes negative for larger houses (with area greater than 142 m 2 ), as indicated by the regression coeffi cient for house area squared, though this U-shaped eff ect is very low in strength (-0.000001). Th e non-linear eff ects of this kind were also confi rmed by Cirman et al (2015), analysing the market for apartments in the wider Ljubljana area, and Romih and Bojnec (2008), based on a nationwide sample of second-hand apartments.…”

“…Th e dependant variable in our analysis is the selling price of a house, which will be explained by employing seven explanatory variables that were proposed by the Surveying and Mapping Authority of the Republic of Slovenia and were proven eff ective by other researchers (Romih and Bojnec, 2008;Cirman, Pahor and Verbič, 2015). Th ese variables are: location, railway proximity, access to gas network, house area, house age, farming lot, and building lot.…”

“…Th ese variables are: location, railway proximity, access to gas network, house area, house age, farming lot, and building lot. Due to increased explanatory power of the model, the dependant variable was defi ned as a natural logarithm of the price in Euros -a transformation that yielded good results in previous studies (Du and Mulley, 2011;Cirman et al, 2015). …”

“…In particular, each model is tested for heteroscedasticity, which introduces bias to the standard errors of regression coeffi cients (White, 1980). Th e problem is very common in housing econometrics due to the complexity of cross-section determinants aff ecting the real estate prices, and was already noted by some other authors analysing the Slovenian real estate market (Cirman et al, 2015;Romih and Bojnec, 2008).…”