Several notions of a R&D paradox can be found in the literature. In the Swedish Paradox version, the emphasis is normally on high and growing levels of business R&D connected to comparatively low GDP growth rates. This paper examines whether this pattern is consistent over time and, more importantly, which sectors drive the aggregate patterns. Based on an investigation of the entire Swedish economy 1985-2001, there is clear evidence that the paradox occurs only in fast-growing manufacturing and service sectors. Fast-growing sectors show an increasing gap between R&D and value-added growth, while the slow-growing sectors do not. This paradox is not interpreted as a sign of failure of the national innovation system, as the largest gap would then be for the slow-growing sectors, failing to transform R&D to economic growth. The gap between R&D and GDP is consistent with the idea of diminishing marginal returns to R&D investment in high-investing sectors. The evidence does not rule out, however, that rendering the innovation system more effective could yield better outcomes. As the findings of a gap are quite consistent over time, it seems fair to conclude that businesses have good reasons for their high R&D investments, despite not being on par with their production growth. (C) 2011 Elsevier B.V. All rights reserved.