In this paper we argue that an oil price change is likely to have greater impact on real GNP in an environment where oil prices have been stable, than in an environment where oil price movement has been frequent and erratic. An oil price shock variable reflecting both the unanticipated component and the time-varying conditional variance of oil price change (forecasts) is constructed and found to be highly significant in explaining economic growth across different sample periods, even when matched against various economic variables and other functions of oil price. We find that positive normalized shocks have a powerful effect on growth while negative normalized shocks do not.

No abstract

We propose a Bayesian stochastic search approach to selecting restrictions for Vector Autoregressive (VAR) models. For this purpose, we develop a Markov Chain Monte Carlo (MCMC) algorithm that visits high posterior probability restrictions on the elements of both the VAR regression coefficients and the error variance matrix. Numerical simulations show that stochastic search based on this algorithm can be effective at both selecting a satisfactory model and improving forecasting performance. To illustrate the potential of our approach, we apply our stochastic search to VAR modelling of inflation transmission from Producer Price Index (PPI) components to the Consumer Price Index (CPI).

No abstract

The costs of state and local pension plans have been a source of fiscal stress in many states and communities. This has led legislatures to consider major changes in these plans. In order to assess the fiscal and staffing consequences of plan changes it is important to develop reliable statistical models of employee retirement behavior. Structural models are valuable in this regard since the proposed reforms often involve fundamental changes in plan design such as a transition from a defined benefit (DB) to a defined contribution (DC) or hybrid plan. In this paper the authors estimate a structural model of teacher retirement behavior using administrative panel data. They show that the Stock-Wise (1990) option value model provides a good fit to the data and predicts well out-of-sample when used to forecast the effect of pension enhancements during the 1990's. The structural model is used to simulate the effect of several DC alternatives to the current DB plan.