Different regions in the same country can have very different MPIs. Overall, the MPI for Kenya (whose regional HDI differences were examined in Chapter 2) is similar to that of India. But in Kenya, the MPI for Nairobi is close to that of Brazil. Central Kenya’s MPI is similar to that of Bolivia. And northeastern Kenya has a worse MPI even than Niger. There were also great inequalities across ethnic groups in Kenya, with 29% of the Embu considered multidimensionally poor, compared with a staggering 96% of the Turkana and Masai peoples. Great inequalities were also found in India, in which indigenous (“tribal”) peoples and low-ranked (“scheduled”) castes were far poorer than people from high-ranking castes. In the Delhi and Kerala regions, just 14 to 16% are MPI poor, but in Jharkhand and Bihar, 77 to 81% are MPI poor. Finally, changes in the MPI over time were examined for three countries: Ghana saw its MPI halved from 0.29 to 0.14; Bangladesh saw its MPI reduced by a more modest 22%; and in Ethiopia, the MPI fell by 16% in the periods studied.