Macroeconomic Factors and Equity Prices: An Empirical Investigation by Using ARDL Approach

Abstract

The relationship between macroeconomic variables and the equity prices has attracted the curiosity of academicians and practitioners since the publication of seminal paper of Chen, et al. (1986). Many empirical studies those tested the relationship reveal that asset pricing theories do not properly identify macroeconomic factors that influence equity prices [Roll and Ross (1980); Fama (1981); Chen, et al. (1986); Hamao (1986); Faff (1988); Chen (1991); Maysami and Koh (2000) and Paul and Mallik (2001)]. In most of these studies, variable selection and empirical analyses is based on economic rationale, financial theory and investors’ intuition. These studies generally apply Eagle and Granger (1987) procedure or Johanson and Jusilieus (1990, 1991) approach in Vector Auto Regressor (VAR) Framework. In Pakistan, Fazal (2006) and Nishat (2001) explored the relationship between macroeconomic factors and equity prices by using Johanson and Jusilieus (1990, 1991) procedure. The present study tests the relationship between macroeconomic variables such as inflation, industrial production, oil prices, short term interest rate, exchange rates, foreign portfolio investment, money supply and equity prices by using Auto Regressive Distributive Lag (ARDL) bounds testing procedure proposed by Pesaran, Shin, and Smith (1996, 2001). The ARDL approach in an errorcorrection setting has been widely applied to examine the impact of macroeconomic factors on economic growth but it is strongly underutilised in the capital market filament of literature. This methodology has a number of advantages over the other models. First, determining the order of integration of macroeconomic factors and equity market returns is not an important issue here because the Pesaran ARDL approach yields consistent estimates of the long-run coefficients that are asymptotically normal irrespective of whether the underlying regressors are I(0) or I(1) and of the extent of cointegration. Secondly, the ARDL approach allows exploring correct dynamic structure while many econometric procedures do not allow to clearly distinguish between long run and short run relationships.