Investment in Greek manufacturing under irreversibility and uncertainty: the message in used capital expenditures

Abstract

This article contributes to the existing literature by showing that uncertainty produces a nonuniform impact to the extent that different types of capital goods exhibit heterogeneous irreversibility, which we define as asset-specific irreversibility. Hence, asset-specific irreversibility is responsible for asymmetries in responses across types of capital goods to uncertainty. We also show that for a given type of capital good, uncertainty produces a variety of responses across sectors, which we define sector-specific irreversibility. In other words, sectoral differences in terms of the ability to substitute a given type of capital with labour, introduce a second-order effect of uncertainty on investment.

Notes

¹ Provided, that the investment opportunity is not of the ‘now-or-never’ type.

² One cannot help spotting a mistreatment, in terms of semantics, of the notion of ‘uncertainty’ in the empirical literature, which usually is equated to ‘risk’. In terms of the ‘Knightian’ (Knight, 1921) distinction, ‘risk’ is used to describe situations in which probabilities are available to guide choice and ‘uncertainty’ to describe situations in which information is too imprecise to be summarized by probabilities. We feel that by employing the Economic Sentiment Indicator our analysis is closer to the concept of Uncertainty rather than Risk.

³ Irreversibility may also be viewed as a ‘lemons’ problem, again in the context of the New Keynesian school, as the outcome of capital market imperfections taking the form of asymmetric information (Stiglitz and Weiss, 1981; Jaffee and Stiglitz, 1990).

⁴ Standard t-tests are very sensitive to deviations from the assumption of normality. The Wilcoxon test is a nonparametric test, and therefore surpasses the problem of specifying a particular distribution for the underlying population (Siegel, 1956; Hollander and Wolfe, 1973; Hettmansperger, 1984).