The literature suggests that cluster firms may undertake both local and nonlocal geographic searches for knowledge that contributes to their product innovation, and that cluster firms must balance their local and nonlocal searches for product innovation. Yet, previous research has seen local and nonlocal searches as one-dimensional, rather than two-dimensional, activities involving search breadth and depth. In this study, we show that local search and nonlocal search are balanced by jointly considering the breadth and depth of geographic search, and that the optimal balance depends on industry dynamism. Using a sample from two industry clusters in China, we find positive relationships between relative local search depth, relative nonlocal search breadth, and the product innovation of cluster firms. Relative local search depth and relative nonlocal search breadth contribute more to product innovation in stable industries than in dynamic industries.
