The aim of this study was to estimate the incidence of T2DM in three population groups: rural, rural-to-urban migrants and urban dwellers. Data from the Peru MIGRANT Study was analysed. The baseline assessment was conducted in 2007-2008 using a single-stage random sample and further follow-up was undertaken in 2015-16. T2DM was defined based on fasting glucose and self-reported diagnosis. Poisson regression models and robust variance to account for cluster effects were used for reporting risk ratios (RR) and 95%CI. At baseline, T2DM prevalence was 8% in urban, 3.6% in rural-to-urban migrants and 1.5% in rural dwellers. After 7.7 (SD: 1.1) years, 6,076 person-years of follow-up, 61 new cases were identified. The incidence rates in the urban, migrant and rural groups were 1.6, 0.9 and 0.5 per 100 person-years, respectively. Relative to rural dwellers, a 4.3-fold higher risk (95%CI: 1.6-11.9) for developing T2DM was found in urban dwellers and 2.7-fold higher (95%CI: 1.1-6.8) in migrants with ≥30 years of urban exposure. Migration and urban exposure were found as significant risk factors for developing T2DM. Within-country migration is a sociodemographic phenomenon occurring worldwide; thus, it is necessary to disentangle the effect of urban exposure on non-healthy habits and T2DM development.