We examine the effects of future renewable expansion in Germany on residual load and renewable surplus generation for policy-relevant scenarios for 2022, 2032 and 2050. We also determine the storage capacities required for taking up renewable surpluses for varying levels of accepted curtailment. Making use of extensive sensitivity analyses, our simulations show that the expansion of variable renewables leads to a strong decrease of the right-hand side of the residual load curve. Renewable surpluses generally have high peaks which only occur in very few hours of the year, whereas overall surplus energy is rather low in most scenarios analyzed. Surpluses increase substantially with growing thermal must-run requirements, decreasing biomass flexibility and decreasing load. On average, most surpluses occur around noon and in spring time. Whereas the energy of single surplus hours is often in the range of existing German pumped hydro capacities, the energy of connected surpluses is substantially larger. Using an optimization model, we find that no additional storage is required in the scenarios for 2022 and 2032 in case of free curtailment. Even restricting curtailment to only 1% of the yearly feed-in of non-dispatchable renewables would render storage investments largely obsolete under the assumption of a flexible system. In contrast, further restrictions of curtailment and a less flexible system would strongly increase storage requirements. In a flexible 2050 scenario, 10 GW of additional storage are optimal even in case of free curtailment due to larger surpluses. Importantly, minor renewable curtailment does not impede achieving the German government's renewable energy targets. We suggest avoiding renewable surpluses in the first place by making thermal generators more flexible. Afterwards, different flexibility options can be used for taking up remaining surpluses, including but not limited to power storage. Curtailment remains as a last resort. Full surplus integration by power storage will never be optimal because of the nature of surpluses shown in this paper. Future research should explore synergies and competition between different flexibility options, while not only covering the wholesale market, but also ancillary services.