Housing markets experience substantial price volatility, short term price change momentum and mean reversion of prices over the long run. Together these features, particularly at their most extreme, produce the classic shape of an asset bubble. In this paper, we review the stylized facts of housing bubbles and discuss theories that can potentially explain events like the boom-bust cycles of the 2000s. Onset of theories assumes rationality and uses idiosyncratic features of the housing market, such as intensive search and short selling constraints, to explain the stylized facts. Cheap credit provides a particularly common rationalization for price booms, but temporary periods of low-interest rates will not explain massive price swings in simple rational models. An incorrectly underpriced default option is needed to explain the formation of rational bubbles. Many non-rational explanations for real estate bubbles exist, but the most promising theories emphasize some form of trend-chasing, which in turn reflects boundedly rational learning.

Between 2000 and 2012, America experienced a great housing convulsion that had all the classic attributes of a real estate bubble. Housing prices rose dramatically and then fell, leaving average real housing prices in 2012 no higher than they were in 2000. Price growth, between 2000 and 2006, was much higher in some places than in others, and the places with the biggest price growth experienced the largest declines. Surprisingly, some of the biggest booms occurred in places like Phoenix and Las Vegas which appear to have few short-run limits on new construction (Nathanson and Zwick, 2014, Gao, 2014, Davidoff, 2013). During the years of the biggest boom—2003, 2004 and 2005—when the change in real housing price growth is regressed on the one-year lag of price growth across metropolitan areas, the coefficient is greater than one. Price growth seemed to build upon itself. This phenomenon represents the more general tendency of price growth to show a strong positive serial correlation at one-year frequencies (Case and Shiller, 1989). There was also a clear pattern of spatial correlation, where a boom that started on the coast seems to have spread to neighboring in-land metropolitan areas (Ferreira and Gyourko, 2012). The U.S. housing cycle that occurred between 2000 and 2012 is extreme but hardly unique. Other countries, such as Ireland and Spain, also experienced housing bubbles and crashes over those years. While Japan’s housing market remained stable after 2000, Japan had experienced its own massive real estate cycle in the 1980s and early 1990s. American history is replete with examples of real estate booms and busts, from the days of the early Republic to the American convulsion of the early 1980s. In summarizing these events, Glaeser (2013) argues that while these events may clearly look like bubbles ex post, even at their height, prices could be reconciled with standard models of real estate evaluation.