India has ambitious renewable energy targets by 2022:100 GW of solar and 60 GW of wind. Both of these technologies are perceived to be more costly than conventional, fossil-fuel, power; and, therefore, require policy support. Using representative, project-level cash flow models, we examine two related questions: First, what would be the cost of policy support under existing federal policies; and, second, what would be the most cost-effective federal policy? We answer these questions by first forecasting the unsubsidized levelized cost of electricity for wind, solar, and the marginal fossil fuel; and then examining the cost of support under existing as well as proposed debt-related policies. We find that wind energy is already competitive with the marginal fossil fuel and, therefore, does not require any policy support. We also find that solar energy will become competitive by 2019; and prior to that the most cost-effective federal policy is provision of reduced-cost long-term debt, which can significantly reduce (by more than 90%) the total cost of support compared to accelerated depreciation, the most cost-effective existing federal policy. (C) 2016 Elsevier Ltd. All rights reserved.