Forecasting 65+ travel : an integration of cohort analysis and travel demand modeling

Abstract

Over the next 30 years, the Boomers will double the 65+ population in the United States and comprise a new generation of older Americans. This study forecasts the aging Boomers' travel. Previous efforts to forecast 65+ travel are lacking in key two respects: they have failed to incorporate generation differences and have forecasted only broad travel characteristics (e.g. vehicle miles traveled). Drawing on the theory of generations, this study investigates empirically whether cohort differences in travel exist between the Boomers and the current 65+ population. It incorporates theoretically motivated cohort variables related to the historical processes of motorization, proxied by registered automobiles per person, and gender role evolution, proxied by labor force participation rates of women. The resulting forecast predicts the aging Boomers' travel demand with respect to activities requiring travel, person miles traveled, usage of transit and non-motorized modes, and trip chaining propensity. Data extracted from the 1977, 1983, 1990, and 1995 National Personal Transportation Surveys (NPTS) are used to estimate discrete and joint discrete/continuous demand models. Multiple imputation is used to impute missing survey data. Iterative proportional fitting is used to simulate future populations for forecasting purposes. Although 65+ travel is predicted to increase across all the modeled travel indicators, the results indicate that the current national forecast of 65+ travel prepared for the National Highway Traffic Safety Administration and the U. S. Department of Health and Human Services may overestimate future demand. The forecasts also suggest that investment in transit could increase 65+ transit usage propensities; opportunities for increasing transit viability are identified.

(cont.) Finally, in the estimated models, the cohort variables are significant, and with the exception of forecasted person-miles, cohort variable inclusion increases forecasted travel. The implication for transportation modeling is that historical location and generation membership affects transportation behavior. The implication for planners is that in preparing for future 65+ transportation needs, studying the current 65+ population is not adequate. The Boomers will comprise a new generation of 65+ with different associated travel needs.