Download This PaperInefficient Bubbles and Efficient Drawdowns in Financial Markets
57 Pages Posted: 16 Jul 2018 Last revised: 17 Apr 2020
Date Written: February 19, 2019
Abstract
At odds with the common “rational expectations” framework for bubbles, economists like Hyman Minsky, Charles Kindleberger and Robert Shiller have documented that irrational behavior, ambiguous information or certain limits to arbitrage are essential drivers for bubble phenomena and financial crises. Following this understanding that asset price bubbles are generated by market failures, we present a framework for explosive semimartingales that is based on the antagonistic combination of (i) an excessive, unstable pre-crash process and (ii) a drawdown starting at some random time. This unifying framework allows one to accommodate and compare many discrete and continuous time bubble models in the literature that feature such market inefficiencies. Moreover, it significantly extends the range of feasible asset price processes during times of financial speculation and frenzy and provides a strong theoretical background for future model design in financial and risk management problem settings. This conception of bubbles also allows us to elucidate the status of rational expectation bubbles, which, by design, suffer from the paradox that a rational market should not allow for misvaluation. While the discrete time case has been extensively discussed in the literature and is most criticized for its failure to comply with rational expectations equilibria, we argue that this carries over to the finite time “strict local martingale”-approach to bubbles. Our framework will simplify and foster interdisciplinary exchange at the intersection of economics and mathematical finance and encourage further research.
Keywords: Financial Bubbles, Financial Crashes, Explosive processes, Bubble decomposition, Strict local martingale approach, Infinite horizon bubbles
JEL Classification: G01, G10, C60
Suggested Citation: Suggested Citation
Swiss Finance Institute ( email )
c/o University of Geneva
40, Bd du Pont-d'Arve
CH-1211 Geneva 4
Switzerland