Climate policy makers aim to accelerate investments in clean energy. A challenge facing investors in liberalized electricity markets is the incompleteness of long-term markets, which can leave investors exposed to unhedged risk. We consider how governments should choose renewable subsidies and carbon taxes in such markets. For this purpose, we develop a new game theoretic, bilevel model that explicitly captures optimal policy choices in anticipation of electricity market behavior. This allows us to endogenize the optimal climate policy decisions of a government maximizing social welfare in a market with risk-averse decision makers. We present an illustrative case study for a power system with a traditional gas technology, variable renewables, storage, and a clean dispatchable technology under demand and gas price uncertainty. We observe that optimal investment tax credits and carbon prices are both higher when long-term markets are missing than when markets are complete. Perhaps surprisingly, investment tax credits are more cost-effective than a carbon tax in some cases where risk markets are missing. This occurs because, by increasing investment in renewables, subsidies can reduce exposure to unhedged risk. A policy mix combining both instruments is the most cost-effective strategy across our experiments.