This article identifies political economy factors that help explain dramatic differences in the pace of child mortality reduction between Tanzania and Uganda from 1995 to 2007. The existing literature largely explains divergence in basic health outcomes with reference to economic variables such as GDP per capita. However, these factors cannot explain recent divergence across African countries with similar levels of GDP per capita, rates of economic growth, and levels of health funding. I argue that institutional and governance divergences between Tanzania and Uganda can be linked directly to differing coverage levels of key child health interventions (especially related to malaria control), and thus to differing child health outcomes. These institutional differences can be explained in part by historical factors, but more relevant causes can be found in recent political events. In Tanzania, there was an unusually effective project of institution building in the health sector, while in Uganda, by contrast, there was a negative political shock to the health system. This was driven by the repatrimonialization of the Ugandan state after President Yoweri Museveni’s decision to eliminate term limits in the 2001–2006 period. This repatrimonialization process reversed previous health sector institutional gains and had particularly negative effects on child health service delivery in Uganda over the period in question.