Essays on capital market integration

Abstract

The thesis comprises three independent essays on capital market integration; focussing on developed financial markets in Europe and GIO industrialised nations. These essays are motivated by a comprehensive theoretical review of the current literature on capital market integration which suggests that further investigation of a number of key issues would be extremely useful. The first essay examines the dynamics of the evolving financial and economic interdependencies between three core European nations (France, Germany and UK) and thirteen other European nations. We employ measures of linear dependence and feedback developed by John Geweke (1982) - JASA, 77, 304-324 - to define periodic integration measures that capture the time varying nature of capital market integration in Europe. Evidence from the tests of capital market integration are analysed in terms of fundamental macroeconomic variables to see whether stock market integration is driven by or dependent on economic convergence. The results suggest that European capital markets are becoming integrated especially since the 1990's. Evidence is found in support of a sfa-ong relationship between our time varying integration measures and some macroeconomic variables indicating an increase in economic convergence. The second essay analyses common asset price behaviour in GIO equity and bond market using an innovative dynamic factor modelling framework. Our methodology combines an observable and a latent variable factor structure and decomposes the total variation in the system into a number of differential effects. Generalised methods of moments (GMM) estimation technique and the Kalman filter are used to derive the decompositions and extract the unobservable factors. The results suggest that GIG equity and bond markets are broadly partitioned on regional lines. However, regional segmentation is more emphatic for the bond markets than for the equity markets. The third essay considers the issue of conditional or time-varying correlations and conditional volatility spillovers across international stock markets. It focuses specifically on conditional sectoral volatility spillovers into the UK stock market and assesses the effects of non-market-wide volatility on UK stock market volatility. The dynamics of volatility emanating from international sectoral portfolios is assessed and their effects on overall UK stock market volatility are discussed. Inter-sectoral volatility transmission between the UK, US and the European markets are also investigated. To extract the time-varying (conditional) correlations between the UK stock markets and the selected US and European sectors and, between the UK sectors and US and European sectors, we rely on both the model by Engle and Kroner (1995) and the dynamic conditional correlation (DCC) model suggested by Engle (2002). The transmission of volatility from the US and European sectors to the UK stock market is assessed in the multivariate generalised autoregressive conditional heteroscedasticity (MVGARCH) model the model of Engle and Kroner (1995). We find substantial evidence of international sectoral volatility spillover into the UK stock market.