Aims/hypothesis The aim of this study was to compare exercise capacity, strength and skeletal muscle perfusion during exercise, and oxidative capacity between South Asians, African Caribbeans and Europeans, and determine what effect ethnic differences in the prevalence of type 2 diabetes has on these functional outcomes. Methods In total, 708 participants (aged [mean±SD] 73 ± 7 years, 56% male) were recruited from the Southall and Brent Revisited (SABRE) study, a UK population-based cohort comprised of Europeans (n = 311) and South Asian (n = 232) and African Caribbean (n = 165) migrants. Measurements of exercise capacity using a 6 min stepper test (6MST), including measurement of oxygen consumption (V O 2) and grip strength, were performed. Skeletal muscle was assessed using near infrared spectroscopy (NIRS); measures included changes in tissue saturation index (ΔTSI%) with exercise and oxidative capacity (muscle oxygen consumption recovery, represented by a time constant [τ]). Analysis was by multiple linear regression. Results When adjusted for age and sex, in South Asians and African Caribbeans, exercise capacity was reduced compared with Europeans (V O 2 [ml min −1 kg −1 ]: β = −1.2 [95% CI-1.9, −0.4], p = 0.002, and β −1.7 [95% CI-2.5, −0.8], p < 0.001, respectively). South Asians had lower and African Caribbeans had higher strength compared with Europeans (strength [kPa]: β = −9 [95% CI-12, −6), p < 0.001, and β = 6 [95% CI 3, 9], p < 0.001, respectively). South Asians had greater decreases in TSI% and longer τ compared with Europeans (ΔTSI% [%]: β = −0.9 [95% CI-1.7, −0.1), p = 0.024; τ [s]: β = 11 [95% CI 3, 18], p = 0.006). Ethnic differences in V O 2 and grip strength remained despite adjustment for type 2 diabetes or HbA 1c (and fat-free mass for grip strength). However, the differences between Europeans and South Asians were no longer statistically significant after adjustment for other possible mediators or confounders (including physical activity, waist-to-hip ratio, cardiovascular disease or hypertension, smoking, haemoglobin levels or β-blocker use). The difference in ΔTSI% between Europeans and South Asians was marginally attenuated after adjustment for type 2 diabetes or HbA 1c and was also no longer statistically significant after adjusting for other confounders; however, τ remained significantly longer in South Asians vs Europeans despite adjustment for all confounders. Conclusions/interpretation Reduced exercise capacity in South Asians and African Caribbeans is unexplained by higher rates of type 2 diabetes. Poorer exercise tolerance in these populations, and impaired muscle function and perfusion in South Asians, may contribute to the higher morbidity burden of UK ethnic minority groups in older age. Keywords Ethnicity. Exercise capacity. Skeletal muscle. Type 2 diabetes Abbreviations ATT Adipose tissue thickness CVD Cardiovascular disease FFM Fat-free mass Hb diff Difference between the oxygenated and deoxygenated haemoglobin signal 6MST 6 min stepper test MVPA Moderate-to-vigorous physic...