In the spirit of questioning the conventional wisdom, here is a thought that occurred to me recently... Many studies seem to show that over long periods of time value stocks outperform growth stocks and small companies outperform larger ones. As a result, many give higher weight to these two classes in their asset allocations.
Now, here is the issue... (let's talk about small stocks, but same logic applies to values ones or really any other subset of the market)
Short version: if small stocks outperform rest of market, they will eventually become 99% of the market
Longer version:
Let's say I split all stocks in the market (set M) into two subsets: small ones (set S) and the rest (set L for "larger"). So M = S + L (where + is union). Now, say people can only invest in two funds: one that invests in all companies in S and another that invests in all companies in L. If we believe that S will outperform L say by 1% on average, that means that over long enough period of time, S will get larger and larger part of M and L will get smaller and smaller chunk of M. So for example after some number of years, S will be 99% of overall market cap M and L will be the remaining 1%...
Do you see how this does not make sense... ? Am I missing anything?
By the way, this same argument is applied by academia to indicate that when you evaluate a value of a company, make sure their terminal growth rate is less than that of the market, because otherwise, it would effectively BECOME the market. I don't recall seeing anyone making this argument for small / value companies as a class though.
Now, here is the issue... (let's talk about small stocks, but same logic applies to values ones or really any other subset of the market)
Short version: if small stocks outperform rest of market, they will eventually become 99% of the market
Longer version:
Let's say I split all stocks in the market (set M) into two subsets: small ones (set S) and the rest (set L for "larger"). So M = S + L (where + is union). Now, say people can only invest in two funds: one that invests in all companies in S and another that invests in all companies in L. If we believe that S will outperform L say by 1% on average, that means that over long enough period of time, S will get larger and larger part of M and L will get smaller and smaller chunk of M. So for example after some number of years, S will be 99% of overall market cap M and L will be the remaining 1%...
Do you see how this does not make sense... ? Am I missing anything?
By the way, this same argument is applied by academia to indicate that when you evaluate a value of a company, make sure their terminal growth rate is less than that of the market, because otherwise, it would effectively BECOME the market. I don't recall seeing anyone making this argument for small / value companies as a class though.