Abstract:
This paper takes advantage of public and non-public patent application
milestones from the IPGOLD database to investigate insider trading pattern
during the patent application process in Australia. Unlike previous research,
this paper uses cumulative change in ownership to measure insider trading
activity. The research also applies di erent centrality concepts in graph the-
ory and network analysis to measure patents' quality. This measure of patent
quality is then used to study whether patent quality motivates insider trad-
ing. The ndings suggest that insiders have insights on patents, presenting
di erent trading behaviours for successful applications versus failed applica-
tions. They only actively trade and accumulate ownership during the appli-
cation process for granted patents. It is further shown that insider trades
around the ling date and the open-to-public-inspection (OPI) date are in-
formative and can help predict application outcome. Informed transactions
concentrate between the ling date and the OPI date, suggesting exploitation
of inside information. Insiders also accumulate more ownership for higher
quality patents but only around earlier milestones. However, these informed
traders show low-level foresight about patent quality. They can utilise appli-
cations information to identify applications with great potentials before such
information is publicly available. Nevertheless, insiders do not show an ability
to predict a patent's prospective in
uence, which is indicated by the number
of accepted claims and forward citations. The ownership accumulation pro-
cess ultimately yields pro ts for insiders. However, they are unlikely to earn
abnormal pro ts in the short-run, indicating that insiders trade with a long-
term perspective and are not found to try pro ting from price
uctuations
around public disclosure events.
