CRIEFF Discussion Paper Number 0104

Cointegrating Relationships between Stock Market Indices and Economic Activity: Evidence from US Data

David McMillan

Abstract
Recent empirical evidence suggests that stock market returns are predictable from a variety of financial and macroeconomic variables. However, with a few exceptions relatively little evidence exists examining the presence of a long-run relationship between these variables. The present paper extends this research by considering whether a cointegrating vector exists between stock market indices and industrial production, inflation, money supply and interest rates. The results suggest positive evidence of cointegration between both S&P500 and DJIA indices and macroeconomic activity variables. The established relationship is positive and significant for industrial production and inflation, negative and significant for long-term interest rates, and negative but insignificant for money supply and short-term interest rates,. These results are consistent with the belief that changes in output which affect expected future cash flows have a positive affect on stock prices, that stocks act as an inflation hedge and that changes in the discount rate have an inverse effect on prices. Variance decompositions show that long-term rates explain a substantial amount of variability in stock prices, whilst short-term rates, industrial production and inflation also have some explanatory power.

JEL Classifications
C22, G12

Keywords
Stock Market Indices, Cointegration, Interest Rates, Variance Decomposition

David G McMillan
University of St Andrews

Back to CRIEFF Discussion Papers list