Aims/hypothesis Potential differences in cardiovascular risk by ethnicity remain uncertain. We evaluated the association of ethnicity with cardiovascular disease (CVD) incidence in a large cohort of people with type 2 diabetes living in Scotland. Methods Data from Scottish Care Information-Diabetes (SCIDiabetes) were linked to Scottish Morbidity Records (SMR01) and National Records of Scotland data for mortality for dates between 2005 and 2011. Of 156,991 people with type 2 diabetes with coded ethnicity, 121,535 (77.4%) had no CVD at baseline (White: 114,461; Multiple Ethnic: 2, 554; Indian: 797; Other Asian: 319; Pakistani: 2,250; Chinese: 387; African-Caribbean: 301 and Other: 466) and were followed up (mean±SD: 4.8±2.3 years) for the development of fatal and non-fatal CVD. Results During follow-up, 16,265 (13.4%) patients developed CVD (ischaemic heart or cerebrovascular diseases). At baseline, Pakistanis were younger and had developed diabetes earlier, had higher HbA 1c and longer duration of diabetes, but had lower BP, BMI, creatinine, proportion of smokers and proportion on antihypertensive therapy than whites. The age and sex adjusted HRs for CVD were HR 1.31 (CI 1.17, 1.47), p<0.001 in Pakistanis and HR 0.66 (CI 0.47, 0.92), p= 0.014 in Chinese compared with whites. Adjusting additionally for an area measure of deprivation, duration of diabetes, conventional CVD and other risk factors, the HR for Pakistanis (HR 1.45 [CI 1.14, 1.85], p=0.002) was significantly higher, and that for Chinese (HR=0.58 [CI 0.24, 1.40], p=0.228) lower, compared with whites. Conclusions/interpretation Compared with whites with type 2 diabetes, those of Pakistani ethnicity in Scotland were at increased risk of CVD, whereas Chinese were at lower risk, with these differences unexplained by known risk factors.Electronic supplementary material The online version of this article (doi:10.1007/s00125-015-3492-0) contains peer-reviewed but unedited supplementary material, which is available to authorised users.