This paper studies how migration affects host economies when firms have market power in both labor and product markets. When firms pay workers below their marginal revenue product and charge prices above marginal cost, migration can reallocate rents across workers, firms, and consumers, so wage changes need not summarize its welfare effects. I examine the Venezuelan exodus to Colombia using confidential firm-level production data that allow me to construct measures of wage markdowns and price markups. A 1 percentage point increase in the local migrant share raises the marginal product of labor by approximately 3 percent and wage markdowns by 5.7 percent, while reducing price markups by 2.4 percent and leaving average wages essentially unchanged. Wage markdowns rise in markets where the shock reduces the value of self-employment, a key outside option for wage workers that disciplines firms' wage-setting power in developing countries, and increase even where no migrant manufacturing workers are measured. The decline in price markups is consistent with demand shifting toward lower-income, more price-sensitive consumers. I develop a general equilibrium model in which firms' labor market power depends on the value of self-employment and poorer consumers are more price-sensitive. I estimate the model on pre-shock cross-sectional moments and validate it against untargeted reduced-form responses to migration. In the model, higher wage markdowns absorb the wage gain from higher productivity. Giving all migrants access to wage employment, formal or informal, nearly eliminates the increase in markdowns and turns a 0.1 percent loss in native real income into a 0.4 percent gain per point of the inflow.