Stock Index Construction – Implications for Trading

I am sufficiently long in the tooth to remember when the only index UK equity investors used as a benchmark was the FT30 share index, in effect the UK’s equivalent of the Dow 30.

The number of indices available, and derivatives based around them, has multiplied hugely since then. But traders need to understand the properties of different indices, and their quirks of behaviour, before risking their capital betting on them.

Most common stock market indices involve capitalisation-weighted arithmetic averages of their constituents. What this means is that the bigger the company in terms of its market capitalisation, the bigger the impact a 1% movement in its price will have on the index.

But most broad market indices now operate a system whereby a constituent’s weighting is modified to reflect the fact that any portion of the issued share capital that is unlikely to be to freely tradable. This is done by eliminating major stakes and cross-shareholdings. The ‘footsie’, S&P, CAC and Dax all work like this.

That’s less important than remembering which indices are the exceptions. The main ones are the Dow and the Nikkei, which are price-weighted averages. Price-weighted simply means that a constituent with a higher share price has more impact than one with a lower one. Since a share price level per se has no axiomatic link to fundamental relative corporate performance, this hardly seems that logical.

Some market indices (the FT30 is one, but they are generally relatively few in number) work on the basis of a geometric average of the indexed change in each constituent. A geometric average multiplies all constituents together and takes the nth root (where ‘n’ is the number of constituents). The drawback of a geometric average is that the rules of mathematics mean that if one constituent were to be valued at zero, the index itself would theoretically have a value of zero.

While this never happens, constituents are promptly replaced if they look like going bust, what it really means is that a failing constituent can distort the index value. The ValueLine index in the US was calculated solely on this basis for many years, but users came to feel the method of calculating it was understating its true performance, ValueLine produced an arithmetically-averaged version as well.

From a trading standpoint, it is worth remembering that capitalisation weighted indices like the ‘footsie’ change their constituents regularly on the basis of set rules. In the ‘footsie’s case this normally happens four times a year, when any constituents below 110th in capitalisation terms are knocked out and any above 90th are automatically included. Between these bands, places change hands in strict order of size.

But the biggest drawback with capitalisation weighted indices is that a handful of companies dominate the index movement. The current largest FTSE100 constituent, HSBC, is capitalised at ten times the size of the 30th (Rolls Royce) and 50 times the 100th constituent (Capital Shopping Centres). Part of the mid-year setback in the FTSE100 can be attributed entirely to the sell-off in BP (currently third largest FTSE100 stock) as a result of the Gulf of Mexico disaster. Bear in mind too that the FTSE100 is by no means the most concentrated index.

The moral is that if you tempted to trade a CFD or an ETF in an unfamiliar index, it pays to be well aware of how the underlying works before you trade. That includes its historic performance and volatility relative to the major world indices, how it is calculated, the degree to which it is concentrated on a handful of stocks, and the relative importance of those stocks in influencing its movements.

Leave a Reply

Your email address will not be published.