Most manipulation experiments simulating global change in tundra were short-term or did not measure plant growth directly. Here, we assessed the growth of three shrubs (Cassiope tetragona, Empetrum hermaphroditum and Betula nana) at a subarctic heath in Abisko (Northern Sweden) after 22 years of warming (passive greenhouses), fertilisation (nutrients addition) and shading (hessian fabric), and compare this to observations from the first decade of treatment. We assessed the growth rate of current-year leaves and apical stem (primary growth) and cambial growth (secondary growth), and integrated growth rates with morphological measurements and species coverage. Primary- and total growth of Cassiope and Empetrum were unaffected by manipulations, whereas growth was substantially reduced under fertilisation and shading (but not warming) for Betula. Overall, shrub height and length tended to increase under fertilisation and warming, whereas branching increased mostly in shaded Cassiope. Morphological changes were coupled to increased secondary growth under fertilisation. The species coverage showed a remarkable increase in graminoids in fertilised plots. Shrub response to fertilisation was positive in the short-term but changed over time, likely because of an increased competition with graminoids. More erected postures and large, canopies (requiring enhanced secondary growth for stem reinforcement) likely compensated for the increased light competition in Empetrum and Cassiope but did not avoid growth reduction in the shade intolerant Betula. The impact of warming and shading on shrub growth was more conservative. The lack of growth enhancement under warming suggests the absence of long-term acclimation for processes limiting biomass production. The lack of negative effects of shading on Cassiope was linked to morphological changes increasing the photosynthetic surface. Overall, tundra shrubs showed developmental plasticity over the longer term. However, such plasticity was associated clearly with growth rate trends only in fertilised plots.