Abstract
This paper examines intentional herding among institutional investors with a particular focus on the technology sector that was the driver of the “New Economy” in the United States during the dot-com bubble of the 1990s. Using data on technology stockholdings of 115 large institutional investors, we test the presence of herding by examining linear dependence and feedback between individual investors’ technology stockholdings and that of the aggregate market. Unlike other models to detect herding, we use Geweke (1982) type causality tests that allow us to disentangle spurious herding from intentional herding via tests of bidirectional and instantaneous causality across portfolio positions in technology stocks. After controlling information based (spurious) herding, our tests show that 38 percent of large institutional investors tend to intentionally herd in technology stocks. The findings support the existing literature that investment decisions by large institutional investors are not only driven by fundamental information, but also by cognitive bias that is characterized by intentional herding.