The impact of monetary policy shocks on financial bubbles: a VAR-based approach

Recent studies have discovered that exogenous and unforecastable policy shifts by the central bank can be disjoined into monetary policy and central bank information components, with distinctive impacts on most macroeconomic variables. This thesis aims to analyze whether policy surprises have different consequences on the formation of financial bubbles. First, a new instrument for the detection of market exuberance is developed by modifying and expanding the GSADF framework of Phillips, Shi, and Yu (2015). Secondly, the series of monetary policy and information shocks from the works of Miranda-Agrippino and Ricco (2021) and Jarociński and Karadi (2020) are incorporated into a Cholesky Vector Autoregression model with policy surprises as internal instruments, based on the equivalence between proxy SVARs and recursive VARs postulated by Plagborg-Møller and Wolf (2021). The dynamic analysis with impulse response functions and forecast error variance decompositions of the baseline model and subsequent robustness checks unveils a new feature of policy surprises: while the two shocks are associated with opposite market reactions, both lead to a reduction in the probability of the development of a bubble in financial markets.