Two forecasting models are developed for forecasting the U.S. manufacturing aggregate fuel and electricity intensities. The models are both simple to apply and capable of identifying the effect of underlying forces of aggregate energy intensity change. The validation of the results provided by these models is performed by comparing these results with those rendered by conventional decomposition techniques based on economic index numbers. The results indicate that the aggregate fuel intensity is expected to decline by 3.2% yr(-1) from the year 2000 to 20 10, of which 1.1% yr(-1) is due to structural effect, i.e. a share of 32.9% of aggregate fuel intensity change. The results also show that in the same period the aggregate electricity intensity is expected to decline at a rate of 1.2% yr(-1), of which 0.6% yr(-1) is due to structural effect, i.e. a share of 46.3% of aggregate electricity intensity change. Copyright (C( 2007 John Wiley & Sons, Ltd.