Untangling demand curves from information effects: evidence from Australian index adjustments

Abstract

We investigate the impact of the Morgan Stanley Capital International change in index calculation on Australian stocks. We are able to differentiate between a downward sloping demand curve hypothesis and the investor awareness hypothesis. Broadly speaking, the results are consistent with investor awareness at the time of the announcements and downward sloping demand curves at the implementation date.

Acknowledgements

The Australian Research Council, SPIRT Grant C79906698, funded the research in this project. We thank Tim Brailsford, Peter Clarkson, Robert Faff and Richard Heaney and an anonymous referee for their helpful and constructive comments on previous versions of this article. We also wish to thank Diana Maldonado for her assistance with the collection of some of the data used in this analysis.

Notes

¹ Although this was the first official announcement of index methodology changes, the market would have been aware of the prospect of changes to MSCI's methodology with the release of a consultation paper on 17 September 2000.

² The criterion that is used for selecting companies includes size, long-term and short-term volume, cross-ownership and available float.

³ The provisional indices were calculated from 31 May 2001, until the completion of the methodology shift on 31 May 2002. During this time, the provisional indices could be used by index funds mandated to replicate the MSCI country indices.

⁴ Wesfarmers acquired Howard Smith Ltd in August 2001. Our analysis proceeds generally with 72 companies. There were only 70 companies for the 31 May 2002 re-balancing following Normandy Mining Limited being acquired by Delta Acquisition LLC and F H Fauldings & Co. becoming part of the Mayne group of companies.

⁵ Companies added to the index were all included with an IF less than one.

⁶ The tables, however, only show results from day −5 to day +5. Full tables are available on request.

⁷ These are the longest nonoverlapping run-up times available given the timing of the announcements.