Common ownership and airlines : evaluating an alternate ownership data source

Starting with the pioneering work of Azar, Schmalz, and Tecu (2018), a rapidly growing body of empirical evidence on the effect of common ownership on market outcomes has emerged. However, testing the robustness of these results to alternative methods and data sources is just beginning. In this paper, we contribute to this growing body of work by comparing results based on two different data sources on institutional ownership: Thomson Reuters Spectrum ("SP") and Thomson Reuters Ownership ("OP"). While SP is used by most researchers in this field, we find that OP has several distinct advantages including broader coverage and more convenient data formatting. We replicate the results of Azar, Schmalz, and Tecu (2018) and show that empirical results change dramatically when using OP instead of SP. We also find evidence that MHHI delta measures using OP data are more volatile than those using the SP data.