Aims/hypothesis FTO harbours the strongest known obesity-susceptibility locus in Europeans. While there is growing evidence for a role for FTO in obesity risk in Asians, its association with type 2 diabetes, independently of BMI, remains inconsistent. To test whether there is an association of the FTO locus with obesity and type 2 diabetes, we conducted a meta-analysis of 32 populations including 96,551 East and South Asians. Methods All studies published on the association between FTO -rs9939609 (or proxy r 2  > 0.98) and BMI, obesity or type 2 diabetes in East or South Asians were invited. Each study group analysed their data according to a standardised analysis plan. Association with type 2 diabetes was also adjusted for BMI. Random-effects meta-analyses were performed to pool all effect sizes. Results The FTO -rs9939609 minor allele increased risk of obesity by 1.25-fold/allele ( p  = 9.0 × 10 −19 ), overweight by 1.13-fold/allele ( p  = 1.0 × 10 −11 ) and type 2 diabetes by 1.15-fold/allele ( p  = 5.5 × 10 −8 ). The association with type 2 diabetes was attenuated after adjustment for BMI (OR 1.10-fold/allele, p  = 6.6 × 10 −5 ). The FTO -rs9939609 minor allele increased BMI by 0.26 kg/m 2 per allele ( p  = 2.8 × 10 −17 ), WHR by 0.003/allele ( p  = 1.2 × 10 −6 ), and body fat percentage by 0.31%/allele ( p  = 0.0005). Associations were similar using dominant models. While the minor allele is less common in East Asians (12–20%) than South Asians (30–33%), the effect of FTO variation on obesity-related traits and type 2 diabetes was similar in the two populations. Conclusions/interpretation FTO is associated with increased risk of obesity and type 2 diabetes, with effect sizes similar in East and South Asians and similar to those observed in Europeans. Furthermore, FTO is also associated with type 2 diabetes independently of BMI.