Reconsidering the role of Tobin's Q Nonlinearities and the adjustment of investment expenditure

Purpose - The purpose of this study is to investigate nonlinearities in the behavior of investment expenditure. Conventional wisdom suggests that Tobin's Q criterion is an important explanation of investment behaviour that bridges the financial and real sides of the economy. However, the empirical evidence in support of Q as a means of explaining aggregate business investment is rather weak. We answer a number of questions about the relationship between investment expenditure and Q. In particular, is the relationship governed by non-linearities? If so, what is the nature of the non-linearities present? Design/methodology/approach - The rationale for paying closer attention to non-linearities is based on the presence of information asymmetries and possible dependence of adjustments on non-linearities with respect to factors such as fixed costs, threshold effects and irreversibility, which are entertained in the investment literature. Using the non-linear vector error-correction model procedure advocated by Hansen and Seo, we show that in the context of the US economy, investment has a long-run relationship with Q that is based on threshold error correction. Findings - There are asymmetries present with respect to error correction or the speed of adjustment towards long-run equilibrium. We find that investment expenditure only responds significantly to long-run disequilibrium from Q during a particular regime. Such a regime is characterised by long-run disequilibrium based on high or rising investment expenditure compared with a relatively weak stock market. Originality/value - The authors provide new insights into the relationship between Tobin's Q and real investment. In contrast to previous work, they find that error correction based on the adjustment of real investment is regime-specific and function of the size of departures from long-run equilibrium. The tests also allow for the identification of periods when error correction has occurred. Not only are these insights significant for future research on financial crises, market volatility and the impact of debt, but for policymaking purposes as well.