This paper models the capital structure decision facing partnerships and tests the implications using panel‐data regression analysis for a sample of real estate limited partnerships. The model shows that if an optimal capital structure exists for non‐taxed firms, it is a function of personal tax effects, costs of financial distress, and substitute tax shields. The empirical tests indicate a positive relationship between leverage and the proportion of real estate assets held, and a negative relationship between leverage and both growth rates and non‐debt tax shields. Furthermore, the findings suggest that changes resulting from the Tax Reform Act of 1986 are positively related to partnership leverage.