OLG Life Cycle Model Transition Paths: Alternate Model Forecast Method

The overlapping generations (OLG) model is an important framework for analyzing any type of question in which age cohorts are affected differently by exogenous shocks. However, as the dimensions and degree of heterogeneity in these models increase, the computational burden imposed by rational expectations solution methods for nonstationary equilibrium transition paths increases exponentially. As a result, these models have been limited in the scope of their use to a restricted set of applications and a relatively small group of researchers. In addition to providing a detailed description of the benchmark rational expectations computational method, this paper presents an alternative method for solving for equilibrium transition paths in OLG life cycle models that is new to this class of model. The key insight is that even naive limited information forecasts within the model produce aggregate time series similar to full information rational expectations time series as long as the naive forecasts are updated each period. We find that our alternate model forecast method reduces computation time by 85 percent, and the approximation error is small.