RDP 2011-06: Does Equity Mispricing Influence Household and Firm Decisions? 2. Related Literature and Approaches

One approach to identifying equity price bubbles and their effects is to take a stand on a specific model that describes the evolution of the economy. This includes a specific model for equity prices, and possibly a model of the process underlying an equity price bubble as well. Once the model has been specified, econometric tests for the presence of bubbles can be undertaken. Reviews that summarise this literature include Gürkaynak (2008), concerning rational bubbles, and Vissing-Jorgensen (2003), covering the literature in behavioural finance.

One advantage of such a structural approach to identification is that the restrictions used are made explicit and can often be tested. However, as noted by Gürkaynak, a common criticism of these hypothesis tests is that they are unable to distinguish between a test for a bubble, and a test of the model assumed as part of the maintained hypothesis. Accordingly, the validity of any results obtained are contingent on the reader accepting the economic model proposed as the correct one. If there was a strong consensus concerning the ‘correct’ or ‘true’ model for the economy and equity pricing this would not be too problematic. However, given a lack of consensus over these issues, it has been difficult for any one structural approach to remain convincing in its ability to detect a bubble.

An alternative approach to identifying bubbles is to use a purely statistical or atheoretical approach. Examples of this approach include Helbling and Terrones (2003), Detken and Smets (2004), and Machado and Sousa (2006). The advantage of atheoretical approaches is that they may be less subject to model misspecification, since they do not rely on any particular assumed model, and can be useful for summarising correlations in the data. However, there remains much scepticism of their ability to identify the effects of bubbles on economic decisions more precisely. This stems from the property that these procedures do not appear well-equipped to distinguish between different sources of movements in equity prices, and thus a boom or bust in equity prices that is identified as a bubble could just as likely reflect improved or worsening fundamentals.

A third approach in the literature, which is closest to this paper, is to use some mix of the structural and statistical approaches. Rather than specifying a tight or unique economic model for fundamentals or bubbles, a weaker set of economic restrictions, consistent with economic theory, is used. These restrictions still provide sufficiently rich information to enable the researcher to get closer to identifying the effects of an equity price bubble, but help to avoid criticisms associated with model specificity. Previous literature in this vein, though not always concerned with identifying the effects of equity market mispricing, includes Cochrane (1994), Lee (1995, 1998), Gallagher (1999), Gallagher and Taylor (2000), Chirinko and Schaller (2001) and Gilchrist et al (2005). This paper contributes to this literature by providing an informative alternative approach to identifying episodes of mispricing in equity markets, and by providing additional evidence on the effects of this mispricing on economic decisions.