Model validation involves the comparison of model forecasts with observed data that are not used in model development to identify model prediction capabilities. In this study, the historical forecasting method of validation is applied to the original version of the Sacramento, California, regional travel demand model (estimated with 1991 data) with observed data obtained in 2000 to test the accuracy of the model over a 9-year period. Two simulations were used to test the model's accuracy and its representation of induced travel. The results indicate that the model's functional forms and parameters overestimate vehicle miles traveled, hours traveled, and hours of delay (by 5.7%, 4.2%, and 17.1%, respectively) and that the model captures about half of estimated induced travel (i.e., elasticity of 0.14 versus 0.22). It appears that if the model were used for air quality analyses in this region, its overestimation of daily vehicle travel would provide a relatively generous margin of error with respect to meeting emissions budgets. However, in environmental impact statements for new highway projects, the model's upward bias would tend to overestimate no-build travel demand and congestion and, thus, the need for the projects. Compared with the no-build alternative, the magnitude of change for the highway alternative would have to be greater than the model error to be considered significantly different.