We compare actual and calibrated values for the foreign exchange risk premium based on the definition in [J. Int. Econ. 32 (1992) 305]. Calibrated values are found from within a dynamic stochastic general equilibrium model of a small open economy consisting of risk averse optimizing agents with unconventional preferences. We find that the equilibrium foreign exchange risk premium is a function of exogenous shocks in the model and is sensitive to assumed attitudes towards risk. Furthermore, various forms of policy uncertainty improve the capacity of the model to generate values closer to those found in the data.

In this paper we use a continuous-time, stochastic, dynamic general equilibrium model to provide estimates of the growth and welfare effects of monetary volatility. Our primary concern is to highlight the long-run consequences of different monetary environments in a small open economy. Using UK-relevant data to set key parameter values in the model, we carry out three policy experiments. We find that (i) eliminating monetary growth shocks and (ii) reducing the infation rate can each generate positive growth and welfare effects, while (iii) reducing the interest rate depresses growth and is welfare deteriorating. However, these results are sensitive to the values set for the risk aversion and intertemporal substitution parameters. Most notably, in some cases, high degrees of risk aversion are suffcient to change the direction of the infuence of volatility on growth and welfare an issue currently challenging the profession.

This paper provides estimates of the welfare cost of volatility attributable to monetary and fiscal policy shocks. It uses a continuous-time stochastic dynamic general equilibrium model based on a recursive utility function that disentangles risk aversion from intertemporal substitution. We find that monetary and fiscal policy shocks may lead to opposite welfare effects: negative for monetary growth shocks, but positive for government expenditure shocks. Furthermore, we find that welfare costs are sensitive to the parameter values chosen for risk aversion and intertemporal substitution, and we conclude that welfare costs are potentially much larger than that found by Lucas, forcing some modification of the policy conclusions associated with Lucas's pioneering work.