As the supply-side reform gets further development, the evolution of electricity demand has become less predictable. Traditional mid/long-term load forecasting models often overlook the supply-side reform factors, making it difficult to meet the requirements on the forecasting accuracy of mid/long-term load under new situation. With the aforementioned background and defect, this paper proposes a novel mid/long-term load forecasting model, named Supply-Side Reform-Extend Solow Regression Model (SSR-ESRM), to adapt to the implementation of the policy. Firstly, the quantitative indicators from scale, structure and efficiency perspective based on the connotation of the structural supply-side reform are constructed. Next, the supply-side reform indicators are introduced into the basic Solow model, extended into SSR-ESRM as underlying point forecast models. Meanwhile, to reflect the uncertainties of this policy, several economic scenarios based on system dynamics model and multiple ex-ante load forecasts are generated. It is shown that SSR-ESRM has higher forecasting accuracy and moderate sensitivity to scenario switching, which can facilitate the flexibility of power system planning and provide useful reference for realizing power grid planning. This work is supported by National Key Basic Research Program of China (973 Program) (No. 2016YFB0900101).