Uncertainty, Investment, and Industry Evolution

Abstract

We study the effects of aggregate and idiosyncratic uncertainty on the entry of firms, total investment, and prices in a competitive industry with irreversible investment. We first use standard dynamic programming methods to determine firms' entry decisions, and we describe the resulting industry equilibrium and its characteristics, emphasizing the effects of different sources of uncertainty. We then show how the conditional distribution of prices can be used as an alternative means of determining and understanding the behavior of firms and the resulting industry equilibrium. Finally, we use four-digit U.S. manufacturing data to examine some implications of the model. JEL Classification Numbers: D92, E22

Innovation or imitation? The effect of spillovers and competitive pressure on firms’ R&D strategy choice

“…Following some previous studies, we expect that unforeseeable demand negatively a¤ects the incentives of …rms to innovate. Especially, it might discourage persistent innovation (see Caballero and Pindyck, 1996;Czarnitzki and Toole, 2008). The asymmetries between …rms are captured by the parameter in the theoretical model.…”

The effects of irreversibility and uncertainty on capital accumulation

“…Caballero's (1991) analysis of the effect of uncertainty on investment concludes that "If this effect is sufficiently strong (i.e., the asymmetry of adjustment costs is large and the negative dependence of the marginal profitability of capital on the level of capital is strong), the investment-uncertainty relationship becomes negative. The irreversibleinvestment arguments analyzed in the literature typically correspond to this case."…”

The Impact of Market Structure and Irreversibility on Investment under Uncertainty: An Empirical Analysis

“…As the level of disaggregation of the data increases, the larger will be the impact of idiosyncratic shocks compared to economy-wide or industry-wide ones, at least for competitive industries. 37 Instead, when studying investment at a more aggregated level, as in this case at industry level, the uncertainty that matters would be the industry-wide one, because the 37 See Caballero and Pindyck (1996) who compare the volatility of marginal profitability of capital for 2-digit industries and the mean of volatility for the 4-digit industries that make up each 2-digit sector, showing that the latter is always higher by a factor of two or three. They explain this with the fact that total uncertainty is made up of idiosyncratic shocks and aggregate uncertainty, and the idiosyncratic shocks are predominant at the more disaggregated level.…”

“…Bernanke (1983) had argued that aggregate and firm-specific shocks could conflict and cancel, altering the impact of uncertainty on investment behaviour. Bertola and Caballero (1994) found empirical evidence of this claim, while Caballero and Pindyck (1996) showed that the effects of uncertainty were usually stronger (if not clearer) at the disaggregated level. The survey by Carruth et al (2000) compare the available empirical evidence, and shows that the aggregate studies of the investmentuncertainty relationship usually find a significant negative sign, while the overall empirical results obtained from the applied works based on disaggregated data are less conclusive.…”

“…So this methodology is questionable. Caballero and Pindyck (1996) apply the same methodology as Pindyck and Solimano (1993) to both 2-digit and 4-digit US industrial data. Their findings however are only weakly supportive of the predictions of the theory because, although they show that increased volatility in the returns to investment leads to higher required rate of return, the size of this effect is very small and not well determined.…”