Speculative bubbles in present-value models: A Bayesian Markov-switching state space approach

Undesired monetary policy effects in a bubbly world

“…Second, we use a less restrictive model for fundamentals by allowing the whole asset price equation to switch across regimes-hence, also its fundamental component. Therefore, as argued in Chan and Santi (2021), the latter also embodies the role of time-varying discount rates and other latent factors driving regimes in fundamentals. Consequently, the residual-namely, our measure for the bubble-is orthogonal to asset price fundamentals and to factors generating shifts in fundamentals.…”

“…The bubble tests developed by Diba and Grossman (1988), Hall et al (1999), andPhillips et al (2015a, b) (among others), indeed, ascertain the existence of bubbles from the explosive dynamics in long time series of asset prices, by using unit root and cointegration analysis on the pricedividend relationship. Other approaches, instead, rely on the present-value literature-initiated by Campbell and Shiller (1988) and followed by many others [Balke and Wohar (2009), Al-Anaswah and Wilfling (2011), and Chan and Santi (2021)]-for which the fundamental price component is derived under risk neutrality as the present discounted value of future dividend flows [Cochrane (2001)], as follows:…”

“…Specifically, a financial bubble is defined as a temporary deviation of a stock price from its fundamental value, namely the present value of future dividend streams [Cochrane (2001)]. Using this convention, Figure 1 shows that most asset price fluctuations are described by periodically growing and collapsing bubble episodes, as also reported by the bubbles detection empirical literature [Driffill and Sola (1998), Hall et al (1999), Brooks and Katsaris (2005), Michaelides et al (2016), and Chan and Santi (2021)]. However, this definition runs into the objection raised by Fama (1991)'s joint hypothesis problem, which questions the ability to reliably detect and evaluate asset price bubbles.…”

A sustainability transition on the move? Evidence based on the disconnect from market fundamentals

“…In line with Binsbergen and Koijen (2010) and Chan and Santi (2021), we assume the growth rate of dividends and gross returns equal their expected values plus an orthogonal shock as in:…”

“…6 The first group applies cointegration and unit root tests in the context of an indirect bubble test looking for explosive behaviour (see Diba and Grossman, 1988a;Phillips et al, 2011Phillips et al, , 2015Homm and Breitung, 2012). The other group applies a Markov-switching-augmented version of the present-value model(see Van Norden and Schaller, 1999;Brooks and Katsaris, 2005;Binsbergen and Koijen, 2010;Al-Anaswah and Wilfling, 2011;Choi et al, 2017;Chan and Santi, 2021). 7 While each of these approaches has its advantages and shortcomings, we follow the latter approach as it provides the advantage of being able to unveil a non-fundamental component and the related estimation uncertainty.…”