In an ideal world, the mechanics of buying and selling - irrespective of the amounts involved – will not influence the price of those securities. But in the real world the positioning of a security is an important factor in its own right, at least in the short term.
A striking example of this was the ‘Flash Crash’ of May 6, 2010, where one benchmark US index lost around 9% of its value in under 10 minutes. Neither classical theory nor fundamental information provided any warning ahead of the event. The post-mortem focused on a wide range of issues from high frequency trading (through changes in market structure), to concentration of ownership.
This question of ‘concentrated ownership’ has gained importance as a few large, passively managed vehicles such as Exchange Traded Funds (ETFs) hold a greater share of many highly traded securities. There is clearly a new type of non-fundamental risk premium emerging because of this growing concentration.
With the growing appetite for passive investment strategies showing no signs of cooling, investors need to start considering the effect of this risk in their traditional and quantitative investment processes, risk models, and factor premiums. Evidence suggests that investors should monitor breadth and concentration of ownership because of the links amongst concentrated ownership of a stock, increased volatility, and lower returns.
Investors should also, where possible, factor in the funding needs and capital flow behaviour of that stock’s owners to assess collective flight risk. Evidence suggests that adopting these measures in quantitative factor and risk models leads to improved outcomes.
Not easy to pin down
Manager positioning is a non-fundamental risk, which means the price movements related to positioning can be totally independent of the securities’ underlying financial dynamics. This is why most companies positioning data is absent from standard reports, and it is hard to obtain from other sources.
Most investors do not report their holdings information and the investors who do report, do so with a low frequency and a significant lag. For example, 13-F filings in the US are done quarterly, typically with a 45-day lag. Moreover, the data is generally at an aggregated at the manager level.
Complete the form to read more of this premium content...