In recent years, renewable energy supply in the U.S. has increased dramatically, among which wind energy has experienced the largest growth. Since the grid is not designed to handle high levels of intermittent generation, involuntary wind curtailment has been implemented commonly in many regional transmission organizations. Transmission congestion is one of the main barriers, which avoids full delivery of the produced wind energy. An obvious but extremely costly solution to this problem is building new transmission lines. Alternatively, transfer capability over the existing transmission system can be enhanced using variable-impedance flexible ac transmission system (FACTS) devices. This paper proposes a two-stage stochastic optimization model for integrating FACTS into power system operation for a system with high levels of wind penetration. The first stage identifies an optimal base dispatch, while the second stage finds the optimal deployment of FACTS flexibilities and operational reserves for each wind scenario. Using this model, the paper studies the impacts of FACTS operation on wind energy integration. The simulation studies on an RTS-96 system verify the effectiveness of the proposed method in reducing the operation cost as well as wind curtailment. The proposed model is also computationally-efficient, as a linear FACTS model is employed.