Mainstream fixed income indices suffer from the same quirk as most equity indices. Click here for more info, but in a nutshell a market capitalization weighted index assigns more weight of a stock in an index based on the relative market capitalization of the stock in question compared to the overall index. So if stock XYZ has a market cap of $5 billion and the index has a total market cap of $100 billion, then stock XYZ has a weighting of 5% in that index. As the price goes up of the stock, so does the market capitalization and hence its weighting in that index. Theretofore price bubbles (irrationally high prices) leads to over exposure to overvalued stocks and underexposure to undervalued stocks (the exact opposite strategy that a rational investor would desire.
The same quirk holds true to fixed income indices which are also market capitalization weighted. For example, a fixed income index would assign the most weight to the largest fixed income issue. In other words your exposures are largest to the companies/entities that have the most debt.
Doesn’t that strike you as counter-intuitive?
Hmmm….that sounds logical to me in the sense that an index is a measure of the relative amounts of debt and interest rates of all debts within the index.
I agree this might not be the best risk management strategy but that is not what indexes are for.
So would it make sense to have an inversely weighted index where you would buy less of the over-valued companies and more of the under-valued companies?
Is this the prequel to a Rafi Fundamental Fixed Income Index announcement?
@Xenko: the trick is determining which is which. I know Preet is into fundamentally-weighted indexing but I for one don’t buy it. With all due respect, it’s just another form of actively managed fund, and for those to work, they must exploit either persistent market inefficiencies or market fluctuations (ie. market timing). Given the weight of evidence behind the efficient-market hypothesis, I’m still waiting for evidence that either approach works.
@Patrick: There wouldn’t be any determining. Just take all the companies in a given index, and then weigh them inversely based on their market caps. Then you would have to test to see if it would give better or worse returns than the standard index. I don’t have access to the data to try and see if the theory would pan out at all, but it would be cool if someone could get all the data necessary to test this.
@Four Pillars – true that was not the original intention of indices, but that is what they are used for NOW! :)
@Xenko – it’s a bit more complicated than that. I’ll explain in the next post…
@Jordan – lol – you’re a sharp cat, they did just launch such an index in the States a few months ago.
@Patrick – there certainly are people who would suggest that it is actively managed, but there are also arguments that it is more transparent of an index than the major cap-weighted indices. Note that many major indices are formed by committee, whereas fundamental indices are screened from all publicly listed stocks measured by FTSE. You might be surprised to know that the formulator of EMH doesn’t believe in it whole-heartedly.
@Xenko – actually, any weighting methodology which severs the link between portfolio weight and price will give the desired result (well, within reason I suppose). Over longer periods of time, it works. Perhaps I’ll start covering this in more detail over the next few weeks.
@Preet: I’d like to see it covered. Can’t wait for future posts. :D