To estimate the mechanical load on the low back in manual materials handling, the Static Strength Prediction Model (SSPM, University of Michigan) is widely used in the occupational field. It requires (for practical reasons) only a small number of input variables (five body segment angles, standing height, total body mass, external load on the hands) on which basis the moment at the lumbo-sacral intervertebral joint (beside other parameters) is computed. The dynamic character of the activities is ignored in the calculations. To evaluate the validity of the SSPM in various situations, lumbar moments in lifting/lowering activities at different lifting techniques and speeds obtained by the SSPM, were compared with those obtained by a more comprehensive dynamic model (DM). An analysis of variance showed significant effects (p = 0.001) of the biomechanical model applied and the lifting speed used on the peak lumbar moment values. No effects of lifting technique were found. The differences in results from the SSPM and DM were dependent on the lifting speed: the SSPM peak lumbar moments were on average 9% (not significant), 21% (significant at p = 0.005) and 42% (p = 0.0001) smaller compared to the DM moments in the slow (mean velocity in a complete lifting/lowering cycle, 0.2 m s-1), normal (0.4 m s-1) and fast (0.8 m s-1) speed condition respectively. The results indicate that the static/dynamic difference between the models is a major source for the different lumbar moments, while other differences between the SSPM and DM are of minor importance.