This paper considers the effects of agglomeration on the production decisions of firms in the high-tech computer cluster. We build upon an alternative definition of the high-tech computer cluster developed by Bardhan et al. (2003) and we exploit a new data source, the National Establishment Time-Series (NETS) Database, to analyze the spatial distribution of firms in this industry. An essential contribution of this research is the recognition that high-tech firms are heterogeneous collections of establishments. We explicitly model the kinship relationships between the headquarters and establishments of these firms and account for their endogenous production technology choices using controls for the spatial and functional configurations of each firm’s establishment locations. The empirical results, from our preferred specification of a random parameters restricted maximum likelihood (REML) production function, are broadly consistent with several recent theoretical models of supply chain management under incomplete contracting (Combes and Duranton (2003), Almazan et al. (2003), and Rotemberg and Saloner (2000)). We find statistically significant and economically meaningful localization effects on high-tech firms’ labor input technology arising from MSA-level proximity to workers in computer services Standard Industrial Classifications (SICs) and establishment-level geographic interactivity. We also find that localization effects have economically significant impacts on the elasticities of other purchased inputs. The channels for these effects are again access to large computer services labor markets and geographically dispersed networks of establishments. Our empirical results indicate that there are few benefits associated with firm locations in labor markets with large numbers of employees in the computer manufacturing sectors. This negative result may reflect the culmination of recent trends in out-sourcing manufactured inputs to distant offshore subsidiaries. Finally, we uncover considerable heterogeneity in the production technologies exploited by firms in the high-tech computer cluster, although in general, the production technology of this industry is characterized by constant returns to scale.