| This dissertation investigates the relationship between location and performance in the United States mutual fund industry. Recent work in economic geography suggests a link between performance of firms and their location relative to other firms in the same industry. Specifically, "clusters" of firms are thought to outperform firms that are isolated. However, much of the literature on economic clusters defines these clusters in an ad hoc manner. This study uses the spatial scan statistic to identify clusters of mutual fund firms. An innovative technique is used to overcome the problem of clusters potentially overlapping one another. Once clusters are identified, their returns are compared to non-clustered firms in the same investment categories. The study finds that some clusters outperform their non-clustered peers in some years, but there is by no means overwhelming evidence that clustering leads to superior performance in general. |
