Estimating Permanent and Transitory Components of Economic Recovery: A Case of Banking Crises

Abstract

This dissertation aims to understand the the general recovery process of real GDP from banking crises from the perspective of permanent and transitory components of output. In chapter 1, we investigate the general long-term effects of systemic banking crises on real GDP using a sample of a number of economies over 1960-2011. Our methodology is to estimate the response of the level of real GDP to a banking crisis by decomposing it into permanent and transitory components using the statistical framework of the Unobserved Components (UC) model (Harvey 1985 and 1989, Clark 1987). Our main empirical findings are summarized as follows. First, we reconfirm that the negative impact of banking crises on output is generally persistent. Second, and more importantly, advanced economies tend to be more adversely affected in magnitude of an output loss, but experience a stronger rebound from recessions. Third, an output loss in countries with well-developed financial markets is largely transitory, while for countries with less developed financial markets, the loss reflects mainly the permanent component. In chapter 2, we investigate the effect of expansionary monetary policy on the recovery process of real GDP in response to a banking crisis using the empirical framework of the UC model. Our sample includes three major economies: US, Japan, and UK. We find that expansionary monetary policy can play an important role in the process of economic recovery. A positive shock of an increase in the money supply reduces more than 50 percent of the transitory losses during the first 4 quarters for the US and Japan, while the impact from monetary policy in the UK is very limited.