Simple and complex identities in high dimensional organizational worlds: dimensional reduction and the emergence of organizational agglomeration

Abstract Organizational clusters are central to how industries evolve, but their origins remain poorly understood. Existing research explains how spatial agglomerations of similar organizations are recognized and labeled and how they influence behavior, yet it rarely addresses how similarity becomes tractable in the first place within high-dimensional organizational worlds. We develop a Schelling agent-based model that formalizes two mechanisms of dimensional reduction: attribute privileging, in which actors selectively foreground a small subset of identity-relevant features, and correlated attribute environments, representing material and technological worlds exhibiting coherent bundles of capabilities. Across four experiments, we show that spatial agglomeration is unlikely without substantial cognitive or environmental compression, that narrowly privileged attributes can stabilize proto-clusters, and that correlated attributes generate robust and durable agglomerations even without focal attention. Together, these results illuminate how the “places and spaces of organizing” become visible long before formal categories or institutions solidify.