The short and long horizon dynamic links between financial variables and the real economy

This paper investigates  the short and long dynamic links between financial variables and real economy. We apply the methodology of Dufour and  Taamouti (2010) who developed  new causality measures which can measure the degree of causality at any horizon h. These causality measures are based on the causality  at horizon h concept introduced in Dufour and Renault (1998) and they consist generalizations of  the one-period causality measures of Geweke (1982). These causality measures are applied to examine whether financial variables cause real economic activity at different horizons and vice versa. We consider five financial variables; these include interest rates, term spreads, stock returns, dividend yields, and exchange rates. The results from applying the methodology of Dufour and  Taamouti (2010) show that financial variables cause real economy several months ahead.  Contrary, conventional causality measures widely used in the literature fail to detect long horizon causality.

Keywords: Short and Long Horizon Causality; Financial variables; Output growth.