On Monday, 23 September 2013, “Kingold Jewelry (NASDAQ:KGJI) was downgraded by analysts at Thomson Reuters/Verus from a buy rating to a hold rating”, according to Zolmax News. The shares closed at $1.73 on Thursday 26 September, up 25% on the day, and up 40.6% from Monday’s close of $1.23.
These kinds of anomalies are pretty rare; living proof of the fallacy of instantaneous market efficiency. But as a wise woman once said, contradictions do not truly exist; “Whenever you think you are facing a contradiction, check your premises. You will find that one of them is wrong” (AR).
In this post, I present a case that alpha can be gleaned from publicly available SEC Form 13-F data. Traditionally, pundits looked at commonalities in institutional top-holdings by dollar amount. Research suggests that these aggregated top holdings among many institutions can be indicative of their “best ideas” (1). That may be all good and well, but common sense indicates that the best leads should come from a good institution’s top holdings per unit of capacity. For my institution, I use RenTec because:
a.) they are quantitative and therefore it may be easier to find commonalities in their holdings; and,
b.) they have consistently delivered exceptional returns.
I believe that their “best ideas” should be those positions in which the position size is largest relative to capacity because a moderately-sized holding for a small float stock is much more indicative of expected risk-reward than a relatively much larger position in a relatively much larger float stock. Additionally, focusing on a single institution (rather than many) allows us to ask the all-important “why” by determining if there are any commonalities in their top holdings. Understanding the “why” might us allow us to move beyond “piggybacking” off of quarterly 13-F data, and understand what drives the decisions of the best in the industry. I argue that if we can deconstruct some of the decision-making criteria, we can use this for finding our own unique source of alpha.