In recent years, the rise and fall of the housing prices have attracted the attentions of general public and researchers. Before 2004 or 2005, many researchers, such as Labonte (2003) and Himmelberg et al. (2005), argued that the rising housing prices could be well explained by the economic fundamentals, and there was little strong evidence to support the occurrence of a housing bubble in the U.S. However, after the dramatic fall of housing prices in 2008, more researchers, such as Mikhed and Zemčík (2009) and Coleman et al. (2008), started to share a similar conclusion that in the U.S., some housing bubble had occurred between 2003 and 2004, and also between 2006 and 2007, although it varied from one metropolitan area to another. Therefore, similar to business cycles, it is difficult to be absolutely certain about the beginning of a (housing) bubble when or before it occurs. This is also why the Fed used to focus on cleaning up the mess after bubbles burst instead of trying to identify and burst a bubble when or before it occurs. At the same time, this is an indication of why it is important to gain a better understanding of bubbles, particularly when one has developed and where one is in the process of its evolution.
As discussed by Kaizoji (2009), the root causes of the recent U.S. housing bubble from 2000 to 2006 may include (i) non-elastic housing supply in (some) metropolitan areas, (ii) declines in the mortgage loan rate and the housing premium due to the massive mortgage credit expansion, and (iii) policy changes from the governments and the Fed. At the same time, there are also other factors that can contribute to the formation of the bubble, such as the speculative behavior of buyers, which is the focus of this paper.