[Author: Mark John Somers, New Jersey Institute of Technology & Rutgers Business School]
Bubbles occur when market factors distort the value of a financial asset such that it greatly exceeds its intrinsic value. At least some of the theory and research in this area has operated from the perspective that asset bubbles are inevitable; that is, bubbles are a characteristic of markets, and under the right conditions, will develop, grow and ultimately burst (cf., Kindleberger & Aliber, 2005). Not surprisingly, a great deal of interest has been expressed in identifying the stages (e.g., life cycles) of asset bubbles not to prevent them, but rather to avoid or mitigate the consequences of crisis and collapse when bubbles deflate.
Information from financial markets generated in vivo as a natural consequence of the formation, growth and unraveling of asset bubbles has provided a rich source of data to model the stages of asset bubbles. Scholars in working in the areas of finance and financial economics have built complex quantitative models designed to assess market risk and the degree to which any given asset might be dangerously overvalued (cf., West, 2002). These models are typically built retrodictively and then used predictively to identify potential and emerging asset bubbles.
This objective is accomplished by identifying extreme levels of market indicators that have been predictive of future steep declines in asset values. As these models are based on underlying financial theory, it is a mistake to consider them “black box” or atheoretical because hypotheses (either implicit or explicit) about the nature of markets are being tested empirically. Further, as data are collected and predictive models are refined, the underlying financial theory is modified accordingly.