Markov switching modelling of interest rate pass-through

The first paper, "Interest rate pass-through and financial crises: do switching regimes matter? The case of Argentina", analyses the dynamic relationship between a money market (interbank) rate and different short-term lending rates by measuring their passthrough. Neither linear single-equation modelling nor linear multi-equation systems capture efficiently this relationship. Several financial crises alter the speed and degree of response to interbank rate shocks. Hence, a Markov switching VAR model shows the pass-through increases considerably for all market interest rates in a high-volatility scenario. The model identifies correctly the periods in which regime shifts occur, and associates them to financial crises. The second paper, "Modelling interest rate pass-through with endogenous switching regimes in Argentina", extends the scope of the Markov switching modelling by including time-varying transition probabilities. Interest rate spreads are used as leading indicators. The model allows devaluation expectations and country risks, (measured by rate spreads) to signal regime switching. Estimation results suggest that the passthrough tends to overshoot with financial instability, but to decrease if that condition is sufficiently large and long-lived. Likewise, results show a quite heterogeneous credit market, with a highly efficient transmission mechanism in the corporate segment, but considerably less in the consumer segment. The final paper, "Regime switching in interest rate pass-through and dynamic bank modelling with risks", builds a theoretical model of dynamic bank optimisation, which provides rationale to a regime-switching behaviour in the interest rate pass-through. It is shown that a regime-switching interbank rate induces a nonlinear behaviour in lending and deposit rates and (by further introducing interbank-alike regime-switching risk premiums) in the pass-through. Thus, the pass-through process is consistent with a nonlinear behaviour even if there are no asymmetric adjustment costs in the response to interbank rate shocks. An empirical application to France and Germany provide results that support these conclusions.EThOS - Electronic Theses Online ServiceUniversity of Warwick. Dept. of EconomicsBanco Central de Reserva del PerúGBUnited Kingdo