Models of Investment under Uncertainty when Shocks are Non-Gaussian. On the Impact of the Policy Uncertainty on Investment

The purpose of the study is to clarify a few possible causes for the delay in new investment projects. It is well known that anticipation of future economic environment and prices affects the current investment decisions of a firm. Gaussian processes are typically used to model returns or prices. The authors believe that in undeveloped markets in transition, which are subject to larger shocks, even aggregate variables deviate significantly from the Brownian motion. The authors construct models for the investment behavior of a firm that make it possible to consider non-gaussian processes as well as analyze how standard models change when we replace a gaussian process by non-gaussian one of the same variance. The authors claim that standard models tend to be rather optimistic. Their model show that in a non-gaussian case, optimal investment would be delayed further, other things being equal. Policy makers are advised to damp fluctuations rather than reduce average volatility as measured by variance. The authors also believe that the prospect of policy changes creates an additional source of uncertainty (a policy uncertainty), thus multiplying the effects of the uncertainty of transition. This means that governments should promise less but make good on their promises immediately, since any delay in the implementation of the promised measures results in additional investment delays.