Phillips curves and natural rates of unemployment provide a poor foundation for analyzing inflation in developing economies. Structuralist alternatives have focused on distributional conflict and cross-sectoral interactions, but if the distributional claims are exogenous, the theory has formal similarities with mainstream analysis, generating a 'natural rate of underemployment'. This paper outlines a modified structuralist model in which historically determined distributional claims eliminate this natural rate of underemployment. Economic development and structural transformation are not blocked by immutable distributional claims, but shocks to relative incomes can produce explosive inflation.