The world of indexes is generally a slow-moving one, with changes to indexes and the markets perception of the ideal characteristics of indexes changing only infrequently.
Recently, Roger Arnott has rocked the indexing world with the introduction of the Research Associates RAFI 1000 index, which is now tradeable in ETF form (ticker: PRF). In short, Arnott contends that market value weighting in indexes is inefficient, and that indexes should be weighted by fundamental factors such as price to book, price to cash flow, and others. Arnott contends that his RAFI 1000 index has dramatically outperformed the S&P 500 for the last 40 years (This is calculated via back testing, of course).
The RAFI has been under attack from various sources, most of whom point out that by weighting the index the way he does, he is simply shifting the index into smaller cap and more towards the "value" end of the spectrum than the S&P 500, and that his outperformance thus comes solely at the expense of risk. An example of such an argument can be found here.
Arnott's rather lame "yeah, you are right, but that doesn't matter" reply to this criticism is here.
My criticism of fundamental weighting and the RAFI 1000 index is different. I contend that it is simply not an "index" at all, but rather an investment strategy masquerading as an index.
What is an index, and what is the purpose of an index? While indexes are used by different people for different purposes, ultimately an index serves to reflect the options that are available to an investor within any asset class. Thus, a total market index represents the entire investment set that someone investing in the total market has to choose from. Thus, one can judge if the person made wise choices. The same holds for any index that is a subset of the total market such as the S&P 500, or the various Russell indexes.
Weighting indexes in any manner other than by market capitalization (perhaps float-adjusted) means that they are no longer indexes at all. Let me illustrate:
Let's say the entire market consists of two stocks: Large stock has a market cap of $900 billion, and small stock has a market weight of $100 billion. Thus, large stock will have a 90% weight in a market cap weighted index and small stock will only have a 10% weight in that index. This is reflective of the investment opportunity that exists for investors. There is $1 trillion of assets in the stock market, with 90% allocated to large stock. Anyone and EVERYONE can track the index by putting 90% of their assets in large stock, and 10% in small stock.
Now let's shift to Arnott's world: Let's say that through his fundamental weighting he winds up with a 50/50 index weighting between the two stocks. Let's say then that everyone decides to become an indexers --- they can't. 50% of the market simply can't invest in small stock at anywhere near the current market price. Thus, the fundamental index IS NOT REPRESENTATIVE OF THE CURRENT INVESTMENT OPPORTUNITY SET, and thus totally and completely worthless as a measurement of investment opportunities and/or investor/manager skill. It is simply an investing strategy, and should be marketed as such.
« Any opinions expressed on the Seeking Alpha sites are those of the individual authors and do not necessarily represent the opinion of SeekingAlpha or its management. »