The present study compared the muscular efficiency in spastic and healthy lower limbs producing the same mechanical work. Sixteen chronic post-stroke hemiparetic and spastic patients and 14 age-matched healthy subjects were submitted to a submaximal stepwise exercise testing on a bicycle ergometer, pedalling with only one lower limb. Net energetic expenditure was computed from oxygen consumption above resting values. Electrical activity of antagonistic muscles in the thigh and in the shank was recorded and co-contraction was defined as the percentage of the pedalling cycle when antagonistic muscles were activated simultaneously. The efficiency was calculated as the ratio between the mechanical work done on the ergometer and the net energetic expenditure. Spasticity was quantitatively evaluated by measuring passive ankle plantar flexor muscle stiffness. The working capacity of the patients' paretic lower limb was very low (<40W). The energy expenditure increased linearly as a function of work intensity, without statistical difference between the patients paretic lower limb (PPL), the patients healthy lower limb (PHL) and the healthy subjects lower limb (HSL). Shank co-contraction was 2.9 times greater in PPL (p<0.05) and 2.3 times greater in PHL (p<0.05) than in HSL. Thigh co-contraction was also 1.8 times greater in PPL than in HSL (p<0.05). The ankle plantar flexor muscle stiffness was statistically greater in PPL than in PHL and HSL (p<0.05). The efficiency was not statistically different between the three groups (p=0.155). In conclusion, the efficiency of work production by paretic and spastic lower limb muscles was normal ( congruent with 20%) despite significant neurological impairments.