Purpose - The purpose of this paper is to re-examine the relationship between real investment and stock prices for the USA for 1960-2005 in view of distinct economic regimes during the 40-year period. Design/methodology/approach - The paper employs simple models of investment, checks for cointegration, and applies the value at risk (VAR) methodology. Findings - First, it was found that during the 1960-1990 period investment and the stock market exhibited a good relationship and shared a common stochastic trend. Second, during the 1990-2005 period this relationship broke down. Finally, extending the model to include the long-term interest rate did not produce significant impacts on or feedbacks from and to either variable. It is concluded that the 1990-2005 period has been distinct from the previous periods in that the stock market did not always abide by the fundamentals such as interest rates and/or investment expenditures. It is thus concluded that the high stock market growth rates of the 1990s have adversely impacted real investment expenditures. Pracical implications - Lack of influence of the real long-term interest rate on either the investment of the stock price equations for the 1990-2005 period. This implies that both investment and the stock market did not “take into account” a fundamental variable, the discount rate, instead they had a run on their own (especially the stock market). Originality/value -The value of the paper is in showing that interest rates and investment expenditures do not always move as economic theory predicts or that economic fundamentals do not always rule.
