Carbon-use efficiency (CUE), the ratio of net primary production (NPP) to gross primary production (GPP), describes the capacity of forests to transfer carbon (C) from the atmosphere to terrestrial biomass. It is widely assumed in many landscape-scale carbon-cycling models that CUE for forests is a constant value of ∼0.5. To achieve a constant CUE, tree respiration must be a constant fraction of canopy photosynthesis. We conducted a literature survey to test the hypothesis that CUE is constant and universal among forest ecosystems. Of the 60 data points obtained from 26 papers published since 1975, more than half reported values of GPP that were not estimated independently from NPP; values of CUE calculated from independent estimates of GPP were greater than those calculated from estimates of GPP derived from NPP. The slope of the relationship between NPP and GPP for all forests was 0.53, but values of CUE varied from 0.23 to 0.83 for different forest types. CUE decreased with increasing age, and a substantial portion of the variation among forest types was caused by differences in stand age. When corrected for age the mean value of CUE was greatest for temperate deciduous forests and lowest for boreal forests. CUE also increased as the ratio of leaf mass-to-total mass increased. Contrary to the assumption of constancy, substantial variation in CUE has been reported in the literature. It may be inappropriate to assume that respiration is a constant fraction of GPP as adhering to this assumption may contribute to incorrect estimates of C cycles. A 20% error in current estimates of CUE used in landscape models (i.e. ranging from 0.4 to 0.6) could misrepresent an amount of C equal to total anthropogenic emissions of CO2 when scaled to the terrestrial biosphere.