Anchor-based multi-view subspace clustering is arousing extensive attention due to its capability to handle large-scale high dimensional datasets. They typically first construct anchor graphs, followed by performing spectral clustering to derive the final clustering results. However, the lack of complementary information from multi-view and local structural information in existing methods have resulted in poor anchor graphs, greatly affecting clustering performance. In this paper, we propose a new multi-view subspace clustering algorithm called Multi-view Subspace clustering via Complementary-enhanced Anchor Graphs (MS-CAG) to tackle this challenge. Specifically, MS-CAG first learns a set of consensus anchors for constructing specific anchor graphs. Meanwhile, we propose Anchor Hilbert Schmidt Independence Criterion (AHSIC) term to penalize dependencies of anchor graphs between views, which promotes the learning of effective complementary information. Furthermore, we design Anchor Local Structure Regularization (ALSR) term to enhance the local structure information of the anchor graph, making it more discriminative. Extensive experiments were conducted on five widely used multi-view datasets illustrate the feasibility and effectiveness of MS-CAG.