Content area
Full Text
1. Introduction
Empirical analysis of auctions of non-distressed properties is a field of research that has received relatively little attention in the real estate literature. One reason may be that the literature is dominated by studies of the USA and other property markets that can be characterized mainly as search markets where negotiation is the commonly used selling mechanism[1]. However, auctions, or auction-like selling environments, have increased in importance also in such markets. For example, Han and Strange (2014) analyzed bidding wars for houses in the US residential market and noted that bidding wars more than tripled as a share of the housing sales between 1995 and 2005[2]. Even though this increase coincided with a booming market, and the share of bidding wars fell after the bust of the housing market, it remains about twice as high as prior to the examined period. Another example of the increased role of auctions as selling mechanism is the UK property market, where specialized property auction houses in the Past decade have experienced a growing interest in selling non-distressed property (both commercial and residential) by the means of auction in a similar fashion as auctions for other durable goods[3].
Although auctions of non-distressed residential properties are most commonly found in booming markets, when demand significantly exceeds supply (Haurin et al., 2013), the practice of sales via auctions has long been the dominant selling mechanism in Sweden, in all phases of the property cycle and for all types of homes, from ordinary to luxurious.
The aim of this paper is to study, empirically, how bidding strategies used by auction participants, as well as the list price strategy used by the seller/broker, affect selling price in the Stockholm condominium market. The data collected for the study concern sales of non-distressed condominium apartments in the inner-city of Stockholm, Sweden. We look specifically into a unique auction data of 629 apartment transactions during January 2010 to December 2011. The data set includes variables that describe the unfolding of the auctions, such as list price, sales price, all bids in the auction and the point of time of each bid, as well as identification codes for each bidder. The data set also contains apartment attributes, such as address, apartment size, number of...