DJ MARKET TALK: FTSE Downweight May Weigh On S&P/ASX 200

Been trying to get to the bottom of the following snippet that was published on September 3...

 0336 GMT [Dow Jones] FTSE Developed Asia Pacific ex-Japan downweight of Australian
equities, resulting from addition of South Korea on Sep. 18, may see up to A$5 billion
worth of funds removed from Australian equities, according to traders and strategists.
Australia's weighting in the index will fall by 9.1 percentage points to 45.82%.


Goldman Sachs JBWere estimates this will generate passive fund outflows of A$600 million
and active fund outflows of A$4 billion. Expects selldown in Australian equities through
Sep. 14-18. 

The announcement of the promotion of South Korea from 'Advanced Emerging' to full blooded 'Developed' was made in September last year.  The effective date for the change is the close of business on 18Sep09 - so index followers presumably are going to reduce their holdings in a manner that will best match the performance of the index to that date - hence the expected selling 14 through to 18 September.

To give you a context, here are FTSE's classification of Asia-Pacific countries:

Developed - Australia, HK-China, Japan, NZ, Singapore
Advanced emerging - Malaysia, South Korea, Taiwan
Secondary emerging - China, India, Indonesia, Pakistan, Phillipines, Thailand

The rather specialised FTSE Developed Asia ex-Japan index had 287 constituents and a market capitalisation of US$1,418bn as at 31Aug09.  Note that 'red chip' shares listed in Hong Kong are also to be reclassified as 'China' effective 18 Sept.

Australia (apparently) makes up 55% of the index or ~A$920bn (which compares to a market capitalisation for the All Ordinaries on the same date of $1,265bn and for the ASX200 of around $1,010bn).

If a 9.1% fall in the index weighting approximates to A$5bn, this implies that FTSE index followers have A$55bn in current holdings in the Australia (or 6% of the index defined universe).  Not sure how the likes of Goldman and AMP have calculated the ~$5bn number - but it seems broadly plausible.

Without obtaining the exact index composition from FTSE, we can assume that it broadly reflects that of S&P200.  The banks and major mining companies will dominate, while property outside the heavyweights will be a rounding error.

Is $5bn a meaningful number then?  Well, daily turnover of the S&P200 is typically in the $3bn to $5bn range.  So the short answer is yes, if the value of shares to be sold is in the order of $5bn it would represent a material amount of weekly volume (call it 20%).

As a cross check consider the volume of secondary issuance by month - note that the market has run higher in recent times against a rising tide of issuance (mind you a lot of this supply has been issued at discounts to prevailing market prices).

In summary, if index followers will be selling circa $5bn shares in the week ending September 18, expect the market to be a little soft at that time...

Postscript:


A couple of further thoughts on the $5bn figure
1) The reweighting will only lead to selling where a fund manager is wholly beholden to the index (eg. $100m fund that is benchmarked to the index).  If a fund manager follows this index amongst others in its universe (eg. a sovereign wealth fund), then the reweighting may simply lead to a reallocation between indices (South Korea drops out of one and reappears in another).  As a somewhat esoteric index in isolation, this must be relevant to some funds - which leads to the next point...
2) I have not come across any funds that are specifically benchmarked to this index.  Typically, a MSCI index is used by fund managers in this region - at least in the retail funds management space.  Probably doesn't say much as it is not as if I have done extensive studies on fund manager benchmarking.
The point is to take the $5bn with a grain of salt.  I have taken Goldman's conclusions at face value.